Sync to trunk.

2025-07-09 03:04:10 -04:00 · 2009-12-26 15:18:40 -05:00 · 2009-12-26 15:18:40 -05:00 · 028136c7ae
commit 028136c7ae
parent a7de858de1 01da2517fc
79 changed files with 15150 additions and 10844 deletions
--- a/Changelog.yaml
+++ b/Changelog.yaml
@ -4,6 +4,116 @@
 # for important features/bug fixes.
 # Also, each release can have new and improved recipes.

+- version: 0.6.30
+  date: 2009-12-26
+
+  new features:
+
+    - title: "Update graphical toolkit to Qt 4.6 for better integration with Windows 7 and OS X Snow Leopard."
+      description: >
+        "The library calibre uses to draw its user interface, Qt, has been updated in all binary builds to 
+        version 4.6. This provides better support for Windows 7 and OS X Snow Leopard,a s well as various
+        speed ups in the user interface and e-book viewer rendering. Note that calibre will still run with
+        Qt 4.5"
+
+    - title: "Device drivers: Support for device specific icons"
+
+    - title: "Add menu options to delete specific formats/covers from the library"
+      tickets: [3509]
+
+    - title: "Metadata dialog: Auto-increment the series number when editing the series and validate the input ISBN based on the check digit"
+      tickets: [4285]
+
+    - title: "Add option to swap author first and last names when reading metadata from a file name"
+
+    - title: "Replace underscores with spaces when reading metadata"
+
+    - title: "Nook driver: Upload covers when sending to device. Also add Output Profile for the Nook"
+
+    - title: "Clicking on row numbers in the book list will now open the book in the viewer"
+      tickets: [4266]
+
+    - title: "Driver for the Boox reader"
+
+
+  bug fixes:
+    - title: "MOBI Metadata reader: Correctly handle non ASCII characters when reading embedded metadata."
+      tickets: [4223]
+
+    - title: "LRF Output: Set category metadata in generated LRF file based on tags"
+      tickets: [4286]
+
+    - title: "News download: Correctly handle URLs with non ASCII characters in them"
+
+    - title: "Fix windows only crash when vieweing MOBI files"
+      tickets: [4259]
+
+    - title: "Remeber selection when sorting and switching between library and device views"
+      tickets: [4279, 4274]
+
+    - title: "Add a retry loop when querying database to workaround intermittent database access problems in windows"
+      tickets: [4264]
+
+    - title: "When adding books, do not add OPF as a separate format"
+
+    - title: "Fix initialize method of plugins not being called"
+
+    - title: "Handle interrupted system calls duriong device detection on OS X"
+      tickets: [4278]
+
+    - title: "EPUB Output: Make splitting to size more accurate by moving it after the workarounds"
+
+    - title: "When adding books via the add books button to the device, restrict to formats supported by device"
+    
+
+  new recipes:
+
+    - title: Cyprus News Live
+      author: kwetal
+
+    - title: Kleine Zeitung
+      author: kwetal
+
+    - title: Business Week Magazine
+      author: Darko Mieltic
+
+    - title: CNET News
+      author: Darko Miletic
+
+    - title: Dilbert
+      author: Darko Miletic
+
+    - title: Le Temps
+      author: Sujata Raman
+
+    - title: Inc magazine
+      author: kwetal
+
+    - title: SME
+      author: kwetal
+
+    - title: Pravda
+      author: kwetal
+
+    - title: Houston Chronicle
+      author: Kovid Goyal
+
+    - title: Strategy and Business
+      author: kwetal
+
+    - title: Watching America
+      author: kwetal
+
+    - title: Aftenposten
+      author: davotibarna
+
+  improved recipes:
+    - Sueddeutsche
+    - Irish Independent
+    - The Straits Times
+    - Harvard Business Review
+    - Wall Street Journal
+
 - version: 0.6.29
  date: 2009-12-18

--- a/resources/images/news/bwmagazine.png
+++ b/resources/images/news/bwmagazine.png
--- a/resources/images/news/cnetnews.png
+++ b/resources/images/news/cnetnews.png
--- a/resources/images/news/columbusdispatch.png
+++ b/resources/images/news/columbusdispatch.png
--- a/resources/images/news/dilbert.png
+++ b/resources/images/news/dilbert.png
--- a/resources/images/news/houston_chronicle-2.png
+++ b/resources/images/news/houston_chronicle-2.png
--- a/resources/recipes/bwmagazine.recipe
+++ b/resources/recipes/bwmagazine.recipe
@ -0,0 +1,64 @@
+
+__license__   = 'GPL v3'
+__copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
+'''
+http://www.businessweek.com/magazine/news/articles/business_news.htm
+'''
+
+from calibre import strftime
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class BWmagazine(BasicNewsRecipe):
+    title                 = 'BusinessWeek Magazine'
+    __author__            = 'Darko Miletic'
+    description           = 'Stay up to date with BusinessWeek magazine articles. Read news on international business, personal finances & the economy in the BusinessWeek online magazine.'
+    publisher             = 'Bloomberg L.P.'
+    category              = 'news, International Business News, current news in international business,international business articles, personal business, business week magazine, business week magazine articles, business week magazine online, business week online magazine'
+    oldest_article        = 10
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    encoding              = 'utf-8'
+    use_embedded_content  = False
+    language              = 'en'
+    INDEX                 = 'http://www.businessweek.com/magazine/news/articles/business_news.htm'
+    cover_url             = 'http://images.businessweek.com/mz/covers/current_120x160.jpg'
+
+
+    conversion_options = {
+                          'comment'          : description
+                        , 'tags'             : category
+                        , 'publisher'        : publisher
+                        , 'language'         : language
+                        }
+
+
+    def parse_index(self):
+        articles = []
+        soup = self.index_to_soup(self.INDEX)
+        ditem = soup.find('div',attrs={'id':'column2'})
+        if ditem:
+            for item in ditem.findAll('h3'):
+                title_prefix = ''
+                description = ''
+                feed_link = item.find('a')
+                if feed_link and feed_link.has_key('href'):
+                    url   = 'http://www.businessweek.com/magazine/' + feed_link['href'].partition('../../')[2]
+                    title = title_prefix + self.tag_to_string(feed_link)
+                    date  = strftime(self.timefmt)
+                    articles.append({
+                                      'title'      :title
+                                     ,'date'       :date
+                                     ,'url'        :url
+                                     ,'description':description
+                                    })
+        return [(soup.head.title.string, articles)]
+
+    keep_only_tags = dict(name='div', attrs={'id':'storyBody'})
+
+    def print_version(self, url):
+        rurl = url.rpartition('?')[0]
+        if rurl == '':
+           rurl = url
+        return rurl.replace('.com/magazine/','.com/print/magazine/')
+
+
--- a/resources/recipes/cnetnews.recipe
+++ b/resources/recipes/cnetnews.recipe
@ -0,0 +1,37 @@
+
+__license__   = 'GPL v3'
+__copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
+'''
+news.cnet.com
+'''
+
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class CnetNews(BasicNewsRecipe):
+    title                 = 'CNET News'
+    __author__            = 'Darko Miletic'
+    description           = 'Tech news and business reports by CNET News. Focused on information technology, core topics include computers, hardware, software, networking, and Internet media.'
+    publisher             = 'CNET'
+    category              = 'news, IT, USA'
+    oldest_article        = 2
+    max_articles_per_feed = 100
+    no_stylesheets        = True
+    encoding              = 'cp1252'
+    use_embedded_content  = False
+    language              = 'en'
+
+    conversion_options = {
+                          'comment'   : description
+                        , 'tags'      : category
+                        , 'publisher' : publisher
+                        , 'language'  : language
+                        }
+
+    remove_tags = [
+                    dict(name='div', attrs={'id':'tweetmemeAndFacebook'})
+                    ,dict(name='ul', attrs={'class':'contentTools'})
+                  ]
+    keep_only_tags = dict(name='div', attrs={'class':'txtWrap'})
+
+    feeds = [(u'News', u'http://news.cnet.com/2547-1_3-0-20.xml')]
+
--- a/resources/recipes/columbusdispatch.recipe
+++ b/resources/recipes/columbusdispatch.recipe
@ -0,0 +1,80 @@
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class ColumbusDispatchRecipe(BasicNewsRecipe):
+    __license__  = 'GPL v3'
+    __author__ = 'kwetal'
+    language = 'en'
+    version = 1
+
+    title = u'The Columbus Dispatch'
+    publisher = u'The Columbus Dispatch'
+    category = u'News, Newspaper'
+    description = u'Daily newspaper from central Ohio'
+
+    use_embedded_content = False
+    remove_empty_feeds = True
+    oldest_article = 1.2
+    max_articles_per_feed = 100
+
+    no_stylesheets = True
+    remove_javascript = True
+    encoding = 'utf-8'
+    # Seems to work best, but YMMV
+    simultaneous_downloads = 2
+
+    # Feeds from http://www.dispatch.com/live/content/rss/index.html
+    feeds = []
+    feeds.append((u'News: Local and state news', u'http://www.dispatch.com/live/static/crt/2_rss_localnews.xml'))
+    feeds.append((u'News: National news', u'http://www.dispatch.com/live/static/crt/2_rss_nationalnews.xml'))
+    feeds.append((u'News: Editorials', u'http://www.dispatch.com/live/static/crt/2_rss_editorials.xml'))
+    feeds.append((u'News: Columnists', u'http://www.dispatch.com/live/static/crt/2_rss_columnists.xml'))
+    feeds.append((u'News: Health news', u'http://www.dispatch.com/live/static/crt/2_rss_health.xml'))
+    feeds.append((u'News: Science news', u'http://www.dispatch.com/live/static/crt/2_rss_science.xml'))
+    feeds.append((u'Sports: OSU football', u'http://www.dispatch.com/live/static/crt/2_rss_osufootball.xml'))
+    feeds.append((u'Sports: OSU men\'s basketball', u'http://www.dispatch.com/live/static/crt/2_rss_osumensbball.xml'))
+    feeds.append((u'Sports: OSU women\'s basketball', u'http://www.dispatch.com/live/static/crt/2_rss_osuwomensbball.xml'))
+    feeds.append((u'Sports: OSU sports', u'http://www.dispatch.com/live/static/crt/2_rss_osusports.xml'))
+    feeds.append((u'Sports: Blue Jackets', u'http://www.dispatch.com/live/static/crt/2_rss_bluejackets.xml'))
+    feeds.append((u'Sports: Crew', u'http://www.dispatch.com/live/static/crt/2_rss_crew.xml'))
+    feeds.append((u'Sports: Clippers', u'http://www.dispatch.com/live/static/crt/2_rss_clippers.xml'))
+    feeds.append((u'Sports: Indians', u'http://www.dispatch.com/live/static/crt/2_rss_indians.xml'))
+    feeds.append((u'Sports: Reds', u'http://www.dispatch.com/live/static/crt/2_rss_reds.xml'))
+    feeds.append((u'Sports: Golf', u'http://www.dispatch.com/live/static/crt/2_rss_golf.xml'))
+    feeds.append((u'Sports: Outdoors', u'http://www.dispatch.com/live/static/crt/2_rss_outdoors.xml'))
+    feeds.append((u'Sports: Cavs/NBA', u'http://www.dispatch.com/live/static/crt/2_rss_cavaliers.xml'))
+    feeds.append((u'Sports: High Schools', u'http://www.dispatch.com/live/static/crt/2_rss_highschools.xml'))
+    feeds.append((u'Sports: Browns', u'http://www.dispatch.com/live/static/crt/2_rss_browns.xml'))
+    feeds.append((u'Sports: Bengals', u'http://www.dispatch.com/live/static/crt/2_rss_bengals.xml'))
+    feeds.append((u'Sports: Auto Racing', u'http://www.dispatch.com/live/static/crt/2_rss_autoracing.xml'))
+    feeds.append((u'Business News', u'http://www.dispatch.com/live/static/crt/2_rss_business.xml'))
+    feeds.append((u'Features: Weekender', u'http://www.dispatch.com/live/static/crt/2_rss_weekender.xml'))
+    feeds.append((u'Features: Life and Arts', u'http://www.dispatch.com/live/static/crt/2_rss_lifearts.xml'))
+    feeds.append((u'Features: Food', u'http://www.dispatch.com/live/static/crt/2_rss_food.xml'))
+    feeds.append((u'Features: NOW! for kids', u'http://www.dispatch.com/live/static/crt/2_rss_now.xml'))
+    feeds.append((u'Features: Travel', u'http://www.dispatch.com/live/static/crt/2_rss_travel.xml'))
+    feeds.append((u'Features: Home and Garden', u'http://www.dispatch.com/live/static/crt/2_rss_homegarden.xml'))
+    feeds.append((u'Features: Faith and Values', u'http://www.dispatch.com/live/static/crt/2_rss_faithvalues.xml'))
+    #feeds.append((u'', u''))
+
+    keep_only_tags = []
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'colhed'}))
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'hed'}))
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'subhed'}))
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'date'}))
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'byline'}))
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'srcline'}))
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'body'}))
+
+    remove_tags = []
+    remove_tags.append(dict(name = 'div', attrs = {'id': 'middle-story-ad-container'}))
+
+    extra_css = '''
+                body {font-family:verdana,arial,helvetica,geneva,sans-serif ;}
+                a {text-decoration: none; color: blue;}
+                div.colhed {font-weight: bold;}
+                div.hed {font-size: xx-large; font-weight: bold; margin-bottom: 0.2em;}
+                div.subhed {font-size: large;}
+                div.date {font-size: x-small; font-style: italic; color: #666666; margin-top: 0.4em; margin-bottom: 0.4em;}
+                div.byline, div.srcline {font-size: small; color: #696969;}
+                '''
+
--- a/resources/recipes/cynewslive.recipe
+++ b/resources/recipes/cynewslive.recipe
@ -0,0 +1,101 @@
+from calibre.web.feeds.news import BasicNewsRecipe
+from datetime import datetime, timedelta
+
+class CyNewsLiveRecipe(BasicNewsRecipe):
+    __license__  = 'GPL v3'
+    __author__ = 'kwetal'
+    language = 'en_CY'
+    version = 1
+
+    title = u'Cyprus News Live'
+    publisher = u'The Cyprus Weekly'
+    category = u'News, Newspaper'
+    description = u'News from Cyprus'
+
+    use_embedded_content = False
+    remove_empty_feeds = True
+    oldest_article = 7
+    max_articles_per_feed = 100
+
+    no_stylesheets = True
+    remove_javascript = True
+
+    pubTime = None
+    minTime = None
+    articleCount = 0
+
+    INDEX = 'http://www.cynewslive.com'
+
+    feeds = []
+    feeds.append(('News: Cyprus', 'http://www.cynewslive.com/main/92,0,0,0-CYPRUS.aspx'))
+    feeds.append(('News: World', 'http://www.cynewslive.com/main/78,0,0,0-UKWORLD.aspx'))
+    feeds.append(('Sport: Football', 'http://www.cynewslive.com/main/82,0,0,0-FOOTBALL.aspx'))
+    feeds.append(('Sport: Rugby', 'http://www.cynewslive.com/main/83,0,0,0-RUGBY.aspx'))
+    feeds.append(('Sport: Cricket', 'http://www.cynewslive.com/main/85,0,0,0-CRICKET.aspx'))
+    feeds.append(('Sport: Tennis', 'http://www.cynewslive.com/main/84,0,0,0-TENNIS.aspx'))
+    feeds.append(('Sport: Other', 'http://www.cynewslive.com/main/86,0,0,0-OTHER.aspx'))
+    feeds.append(('Business: Local', 'http://www.cynewslive.com/main/100,0,0,0-LOCAL.aspx'))
+    feeds.append(('Business: Foreign', 'http://www.cynewslive.com/main/101,0,0,0-FOREIGN.aspx'))
+    feeds.append(('Environment', 'http://www.cynewslive.com/main/93,0,0,0-ENVIRONMENT.aspx'))
+    feeds.append(('Culture', 'http://www.cynewslive.com/main/208,0,0,0-CULTURE.aspx'))
+
+    keep_only_tags = []
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'ArticleCategories'}))
+
+    extra_css = '''
+                body{font-family:verdana,arial,helvetica,geneva,sans-serif ;}
+                '''
+
+    def parse_index(self):
+        answer = []
+        for feed in self.feeds:
+            self.articleCount = 0
+            articles = []
+            soup = self.index_to_soup(feed[1])
+
+            table = soup.find('table', attrs = {'id': 'ctl00_cp_ctl01_listp'})
+            if table:
+                self.pubTime = datetime.now()
+                self.minTime = self.pubTime - timedelta(days = self.oldest_article)
+
+                self.find_articles(table, articles)
+
+            answer.append((feed[0], articles))
+
+        return answer
+
+    def postprocess_html(self, soup, first):
+        for el in soup.findAll(attrs = {'style': True}):
+            del el['style']
+
+        for el in soup.findAll('font'):
+            el.name = 'div'
+            for attr, value in el:
+                del el[attr]
+
+        return soup
+
+    def find_articles(self, table, articles):
+        for div in table.findAll('div', attrs = {'class': 'ListArticle'}):
+            el = div.find('div', attrs = {'class': 'ListArticle_T'})
+            title = self.tag_to_string(el.a)
+            url = self.INDEX + el.a['href']
+
+            description = self.tag_to_string(div.find('div', attrs = {'class': 'ListArticle_BODY300'}))
+
+            el = div.find('div', attrs = {'class': 'ListArticle_D'})
+            if el:
+                dateParts = self.tag_to_string(el).split(' ')
+                monthNames = {'January': 1, 'February': 2, 'March': 3, 'April': 4, 'May': 5, 'June': 6,
+                              'July': 7, 'August': 8, 'September': 9, 'October': 10, 'November': 11,
+                              'December': 12}
+                timeParts = dateParts[3].split(':')
+                self.pubTime = datetime(year = int(dateParts[2]), month = int(monthNames[dateParts[1]]),
+                                        day = int(dateParts[0]), hour = int(timeParts[0]),
+                                        minute = int(timeParts[1]))
+
+            if self.pubTime >= self.minTime and self.articleCount <= self.max_articles_per_feed:
+                articles.append({'title': title, 'date': self.pubTime, 'url': url, 'description': description})
+                self.articleCount += 1
+            else:
+                return
--- a/resources/recipes/dilbert.recipe
+++ b/resources/recipes/dilbert.recipe
@ -0,0 +1,41 @@
+__license__   = 'GPL v3'
+__copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
+'''
+http://www.dilbert.com
+'''
+
+from calibre.web.feeds.recipes import BasicNewsRecipe
+
+class DosisDiarias(BasicNewsRecipe):
+    title                  = 'Dilbert'
+    __author__             = 'Darko Miletic'
+    description            = 'Dilbert'
+    oldest_article         = 5
+    max_articles_per_feed  = 100
+    no_stylesheets         = True
+    use_embedded_content   = True
+    encoding               = 'utf-8'
+    publisher              = 'UNITED FEATURE SYNDICATE, INC.'
+    category               = 'comic'
+    language               = 'en'
+
+    conversion_options = {
+                             'comments'        : description
+                            ,'tags'            : category
+                            ,'language'        : language
+                            ,'publisher'       : publisher
+                         }
+
+    feeds = [(u'Dilbert', u'http://feeds.dilbert.com/DilbertDailyStrip' )]
+
+    def get_article_url(self, article):
+        return article.get('feedburner_origlink', None)
+
+    def preprocess_html(self, soup):
+       for tag in soup.findAll(name='a'):
+           if tag['href'].find('http://feedads') >= 0:
+              tag.extract()
+       return soup
+
+
+
--- a/resources/recipes/hbr.recipe
+++ b/resources/recipes/hbr.recipe
@ -11,17 +11,15 @@ class HBR(BasicNewsRecipe):
    language = 'en'
    no_stylesheets = True

-    LOGIN_URL = 'http://hbr.harvardbusiness.org/login?request_url=/'
-    INDEX = 'http://hbr.harvardbusiness.org/current'
-
-    keep_only_tags = [dict(name='div', id='content')]
-    remove_tags = [
-            dict(id=['articleDate', 'subscriptionModule', 'errorArea',
-                'feedbackForm', 'relatedModule', 'articleSecondaryModule',
-                'contentRight', 'summaryLink']),
-            dict(name='form'),
-            ]
+    LOGIN_URL = 'http://hbr.org/login?request_url=/'
+    INDEX = 'http://hbr.org/current'

+    keep_only_tags = [dict(name='div', id='pageContainer')]
+    remove_tags = [dict(id=['mastheadContainer', 'magazineHeadline',
+        'articleToolbarTopRD', 'pageRightSubColumn', 'pageRightColumn',
+        'todayOnHBRListWidget', 'mostWidget', 'keepUpWithHBR',
+        'mailingListTout', 'partnerCenter', 'pageFooter']),
+        dict(name='iframe')]
    extra_css = '''
                a {font-family:Georgia,"Times New Roman",Times,serif; font-style:italic; color:#000000; }
                .article{font-family:Georgia,"Times New Roman",Times,serif; font-size: xx-small;}
@ -34,14 +32,14 @@ class HBR(BasicNewsRecipe):
    def get_browser(self):
        br = BasicNewsRecipe.get_browser(self)
        br.open(self.LOGIN_URL)
-        br.select_form(nr=0)
-        br['ssousername'] = self.username
-        br['password'] = self.password
+        br.select_form(name='signInForm')
+        br['signInForm:username'] = self.username
+        br['signInForm:password'] = self.password
        raw = br.submit().read()
        if 'My Account' not in raw:
            raise Exception('Failed to login, are you sure your username and password are correct?')
        self.logout_url = None
-        link = br.find_link(text='(sign out)')
+        link = br.find_link(text='Sign out')
        if link:
            self.logout_url = link.absolute_url
        return br
@ -54,56 +52,70 @@ class HBR(BasicNewsRecipe):
        if url.endswith('/ar/1'):
            return url[:-1]+'pr'

-    def get_features(self, soup):
-        div = soup.find('div', id='issueFeatures')
-        for li in div.findAll('li'):
-            a = li.find('a', href=True)
-            url = 'http://hbr.harvardbusiness.org'+a['href']
-            url = self.map_url(url)
-            if not url:
-                continue
-            title = self.tag_to_string(a)
-            p = li.find('p')
-            desc = ''
-            if p is not None:
-                desc = self.tag_to_string(p)
-            yield {'title':title, 'url':url, 'description':desc}

-    def get_departments(self, soup):
-        div = soup.find('div', id='issueDepartmentsContent')
-        for h4 in div.findAll('h4'):
-            feed = self.tag_to_string(h4)
-            articles = []
-            ul = h4.findNextSibling('ul')
-            for li in ul.findAll('li'):
-                a = li.find('a', href=True)
-                url = 'http://hbr.harvardbusiness.org'+a['href']
-                url = self.map_url(url)
-                if not url:
-                    continue
-                title = self.tag_to_string(a)
-                p = li.find('p')
-                desc = ''
-                if p is not None:
-                    desc = self.tag_to_string(p)
-                articles.append({'title':title, 'url':url, 'description':desc})
-            yield [feed, articles]
+    def hbr_get_toc(self):
+        soup = self.index_to_soup(self.INDEX)
+        url = soup.find('a', text=lambda t:'Full Table of Contents' in t).parent.get('href')
+        return self.index_to_soup('http://hbr.org'+url)
+
+    def hbr_parse_section(self, container, feeds):
+        current_section = None
+        current_articles = []
+        for x in container.findAll(name=['li', 'h3', 'h4']):
+            if x.name in ['h3', 'h4'] and not x.findAll(True):
+                if current_section and current_articles:
+                    feeds.append((current_section, current_articles))
+                current_section = self.tag_to_string(x)
+                current_articles = []
+                self.log('\tFound section:', current_section)
+            if x.name == 'li':
+                a = x.find('a', href=True)
+                if a is not None:
+                    title = self.tag_to_string(a)
+                    url = a.get('href')
+                    if '/ar/' not in url:
+                        continue
+                    if url.startswith('/'):
+                        url = 'http://hbr.org'+url
+                    url = self.map_url(url)
+                    p = x.find('p')
+                    desc = ''
+                    if p is not None:
+                        desc = self.tag_to_string(p)
+                    if not title or not url:
+                        continue
+                    self.log('\t\tFound article:', title)
+                    self.log('\t\t\t', url)
+                    self.log('\t\t\t', desc)
+                    current_articles.append({'title':title, 'url':url,
+                        'description':desc, 'date':''})
+        if current_section and current_articles:
+            feeds.append((current_section, current_articles))
+
+
+
+    def hbr_parse_toc(self, soup):
+        feeds = []
+        features = soup.find(id='issueFeaturesContent')
+        self.hbr_parse_section(features, feeds)
+        departments = soup.find(id='issueDepartments')
+        self.hbr_parse_section(departments, feeds)
+        return feeds
+

    def parse_index(self):
-        soup = self.index_to_soup(self.INDEX)
-        feeds = []
-        feeds.append(('Features', list(self.get_features(soup))))
-        feeds.extend(self.get_departments(soup))
+        soup = self.hbr_get_toc()
+        feeds = self.hbr_parse_toc(soup)
        return feeds

    def get_cover_url(self):
        cover_url = None
-        index = 'http://hbr.harvardbusiness.org/current'
+        index = 'http://hbr.org/current'
        soup = self.index_to_soup(index)
-        link_item = soup.find('img', alt=re.compile("HBR Cover Image"), src=True)
+        link_item = soup.find('img', alt=re.compile("Current Issue"), src=True)

        if link_item:
-           cover_url = 'http://hbr.harvardbusiness.org' + link_item['src']
+           cover_url = 'http://hbr.org' + link_item['src']

        return cover_url

--- a/resources/recipes/houston_chronicle.recipe
+++ b/resources/recipes/houston_chronicle.recipe
@ -0,0 +1,66 @@
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class HoustonChronicle(BasicNewsRecipe):
+
+    title          = u'The Houston Chronicle'
+    description    = 'News from Houston, Texas'
+    __author__	   = 'Kovid Goyal'
+    language       = 'US'
+    timefmt        = ' [%a, %d %b, %Y]'
+    no_stylesheets = True
+
+    keep_only_tags = [dict(id=['story-head', 'story'])]
+    remove_tags = [dict(id=['share-module', 'resource-box',
+        'resource-box-header'])]
+
+    def parse_index(self):
+        soup = self.index_to_soup('http://www.chron.com/news/')
+        container = soup.find('table', attrs={'class':'body-columns'})
+
+        feeds = []
+        current_section = 'Top Stories'
+        current_articles = []
+
+        self.log('\tFound section:', current_section)
+
+        for div in container.findAll('div'):
+            if div.get('class', None) == 'module-mast':
+                t = self.tag_to_string(div).replace(u'\xbb', '').strip()
+                if t and 'interactives' not in t:
+                    if current_section and current_articles:
+                        feeds.append((current_section, current_articles))
+                    current_section = t
+                    current_articles = []
+                    self.log('\tFound section:', current_section)
+            elif div.get('storyid', False):
+                a = div.find('a', href=True)
+                if a:
+                    title = self.tag_to_string(a)
+                    url = a.get('href')
+                    if title and url:
+                        if url.startswith('/'):
+                            url = 'http://www.chron.com'+url
+                        self.log('\t\tFound article:', title)
+                        self.log('\t\t\t', url)
+                        current_articles.append({'title':title, 'url':url,
+                            'date':'', 'description':''})
+            elif div.get('class', None) == 'columnbox' and \
+                    'special' in current_section.lower():
+                a = div.find('a')
+                if a:
+                    title = self.tag_to_string(a)
+                    url = a.get('href')
+                    if title and url:
+                        if not url.startswith('/'): continue
+                        url = 'http://www.chron.com'+url
+                        self.log('\t\tFound article:', title)
+                        self.log('\t\t\t', url)
+                        a.extract()
+                        desc = self.tag_to_string(div)
+                        current_articles.append({'title':title, 'url':url,
+                            'date':'', 'description':desc})
+
+        if current_section and current_articles:
+            feeds.append((current_section, current_articles))
+        return feeds
+
--- a/resources/recipes/inc.recipe
+++ b/resources/recipes/inc.recipe
@ -0,0 +1,71 @@
+from calibre.web.feeds.news import BasicNewsRecipe
+import re
+
+class IncMagazineRecipe(BasicNewsRecipe):
+    __license__  = 'GPL v3'
+    __author__ = 'kwetal'
+    language = 'en'
+    version = 1
+
+    title = u'Inc Magazine'
+    publisher = u'Mansueto Ventures LLC'
+    category = u'News, Business'
+    description = u'Handbook of the American Entrepeneur'
+
+    use_embedded_content = False
+    remove_empty_feeds = True
+
+    no_stylesheets = True
+    remove_javascript = True
+
+    INDEX = 'http://www.inc.com/magazine'
+
+    remove_tags = []
+    remove_tags.append(dict(name = 'div', attrs = {'id' : 'advt'}))
+
+    extra_css = '''
+                body{font-family:verdana,arial,helvetica,geneva,sans-serif ;}
+                div#deck {font-weight: bold;}
+                div.byline {font-size: x-small; color: #696969; margin-top: 0.4em;}
+                '''
+
+    def parse_index(self):
+        soup = self.index_to_soup(self.INDEX)
+
+        url = self.browser.geturl()
+        date = url.rpartition('/')[0].rpartition('/')[2]
+        self.title = self.title + ' ' + date[4:6] + ', ' + date[0:4]
+
+        answer = []
+
+        for feature in soup.findAll('div', attrs = {'class': re.compile('magazinesection.*')}):
+            h2 = feature.find('h2')
+            if h2:
+                feedTitle = self.tag_to_string(h2)
+            else:
+                img = feature.find('img', attrs = {'class': 'howtohead'})
+                if img:
+                    feedTitle = img['alt']
+                else:
+                    feedTitle = 'Unknown Feature'
+
+            articles = []
+            for div in feature.findAll('div', attrs = {'class': re.compile('article.*|column.*')}):
+                h3 = div.find('h3')
+                title = self.tag_to_string(h3)
+                href = h3.a['href'].replace('.html', '_Printer_Friendly.html')
+                p = div.find('p', attrs = {'class': 'deck'})
+                description = self.tag_to_string(p)
+
+                articles.append({'title': title, 'date': u'', 'url': href, 'description': description})
+
+            answer.append((feedTitle, articles))
+
+        return answer
+
+    def preprocess_html(self, soup):
+        img = soup.find('img', attrs = {'src': 'http://images.inc.com/nav/lofi_logo.gif'})
+        if img:
+            img.parent.extract()
+
+        return soup
--- a/resources/recipes/irish_independent.recipe
+++ b/resources/recipes/irish_independent.recipe
@ -15,6 +15,7 @@ class IrishIndependent(BasicNewsRecipe):
    max_articles_per_feed = 100
    remove_tags_before = dict(id='article')
    remove_tags_after  = [dict(name='div', attrs={'class':'toolsBottom'})]
+    no_stylesheets = True
    remove_tags    = [
 	        dict(name='div',attrs={'class':'toolsBottom'}),
 	        dict(name='div',attrs={'class':'toolsTop'}),
--- a/resources/recipes/kleinezeitung.recipe
+++ b/resources/recipes/kleinezeitung.recipe
@ -0,0 +1,53 @@
+from calibre.web.feeds.news import BasicNewsRecipe
+import re
+
+class KleineZeitungRecipe(BasicNewsRecipe):
+    __license__  = 'GPL v3'
+    __author__ = 'kwetal'
+    language = 'de_AT'
+    version = 1
+
+    title = u'Kleine Zeitung'
+    publisher = u'Kleine Zeitung GmbH & Co KG'
+    category = u'News, Newspaper'
+    description = u'Nachrichten aus \u00D6sterreich'
+
+    use_embedded_content = False
+    remove_empty_feeds = True
+    oldest_article = 2
+    max_articles_per_feed = 100
+
+    no_stylesheets = True
+    remove_javascript = True
+
+    # Feeds from http://www.kleinezeitung.at/allgemein/multimedia/102434/wichtige-news-immer-sofort-ueber-rss-feed-abrufen.story
+    feeds = []
+    feeds.append((u'Chronik', u'http://www.kleinezeitung.at/klon/rss/nachrichten'))
+    feeds.append((u'Wirtschaft', u'http://www.kleinezeitung.at/klon/rss/wirtschaft'))
+    feeds.append((u'Leute', u'http://www.kleinezeitung.at/klon/rss/leute'))
+    feeds.append((u'Sport', u'http://www.kleinezeitung.at/klon/rss/sport'))
+    feeds.append((u'Nachrichten aus der Steiermark', u'http://www.kleinezeitung.at/klon/rss/steiermark'))
+    feeds.append((u'Nachrichten aus Kaernten', u'http://www.kleinezeitung.at/klon/rss/kaernten'))
+    feeds.append((u'Multimedia-News', u'http://www.kleinezeitung.at/klon/rss/multimedia'))
+    feeds.append((u'Kino, Events & Tickets', u'http://www.kleinezeitung.at/klon/rss/events'))
+
+    keep_only_tags = []
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'article_body'}))
+
+    remove_tags = []
+    remove_tags.append(dict(name = 'a', attrs = {'id': 'comment_count'}))
+    remove_tags.append(dict(name = 'div', attrs = {'class': re.compile('adv[0-9]+')}))
+    remove_tags.append(dict(name = 'div', attrs = {'class': 'art_info'}))
+    remove_tags.append(dict(name = 'div', attrs = {'id': re.compile('grafikoverlay_.*')}))
+    remove_tags.append(dict(name = 'a', attrs = {'class': 'zoom'}))
+
+    extra_css = '''
+                body {font-family:verdana,arial,helvetica,geneva,sans-serif ;}
+                h1 {text-align: left;}
+                span {margin-left: 0.1em; margin-right: 0.1em;}
+                span.update {font-size: x-small; color: #666666}
+                span.update strong {font-weight: normal;}
+                p.intro {font-size: large;}
+                div.art_foto_big, div.art_foto {font-size: xx-small; color: #696969; margin-bottom: 0.5em;}
+                div.art_foto_big span.src {float: right;}
+                '''
--- a/resources/recipes/le_temps.recipe
+++ b/resources/recipes/le_temps.recipe
@ -14,75 +14,77 @@ class LeTemps(BasicNewsRecipe):
     title          = u'Le Temps'
     oldest_article = 7
     max_articles_per_feed = 100
+     __author__ = 'Sujata Raman'
     no_stylesheets = True
-     remove_tags    = [dict(name='div', attrs={'id':'footer'})]
-     remove_tags    = [dict(name='div', attrs={'class':'box links'})]
-     remove_tags    = [dict(name='script')]
-     extra_css      = '''.heading {font-size: 13px; line-height: 15px;
- margin: 20px 0;} \n h2 {font-size: 24px; line-height: 25px; margin-bottom:
- 14px;} \n .author {font-size: 11px; margin: 0 0 5px 0;} \n .lead {font-
- weight: 700; margin: 10px 0;} \n p {margin: 0 0 10px 0;}'''
+     remove_javascript = True
+     recursions = 1
+     encoding = 'UTF-8'
+     match_regexps = [r'http://www.letemps.ch/Page/Uuid/[-0-9a-f]+\|[1-9]']
+     lang = 'fr'
+
+     keep_only_tags = [dict(name='div', attrs={'id':'content'}),
+                        dict(name='div', attrs={'class':'story'})
+                      ]
+     remove_tags    = [dict(name='div', attrs={'id':['footer','sub']}),
+                       dict(name='div', attrs={'class':['box additional','box function','right','box links','follow']})]
+
+
+     extra_css      = '''h1{font-family:"Georgia","Times New Roman",Times,serif;font-size:large;}
+                         .headline{font-family:"Georgia","Times New Roman",Times,serif;font-size:large;color:#990000;}
+                         .summary_gal{color:#777777;font-family:"Georgia","Times New Roman",Times,serif;font-size:x-small;}
+                         #capt{color:#1B1B1B;font-family:"Georgia","Times New Roman",Times,serif;font-size:x-small;}
+                         #content{font-family:"Lucida Grande","Lucida Sans Unicode",Arial,Verdana,sans-serif;}
+                         h2 {font-size: 24px; line-height: 25px; margin-bottom: 14px; text-transform:uppercase;}
+                         .author {font-size:x-small; margin: 0 0 5px 0; color:#797971; font-family:"Lucida Grande","Lucida Sans Unicode",Arial,Verdana,sans-serif;}
+                         .lead {font-family:"Lucida Grande","Lucida Sans Unicode",Arial,Verdana,sans-serif;font-weight: bold; margin: 10px 0;font-size:small;}
+                         p {margin: 0 0 10px 0;}
+                         h3{font-size:small;font-weight:bold;}
+                         .heading{color:#940026;font-size:x-small;}
+                         .description{font-size:x-small;font-family:"Lucida Grande","Lucida Sans Unicode",Arial,Verdana,sans-serif;color:#797971; }
+                         a {color:#1B1B1B; font-size:small;}
+                         .linkbox{font-size:x-small;color:#1B1B1B;font-family:"Lucida Grande","Lucida Sans Unicode",Arial,Verdana,sans-serif;} '''

     feeds          = [
-                              ('Actualité',
- 'http://www.letemps.ch/rss/site/'),
-                              ('Monde',
- 'http://www.letemps.ch/rss/site/actualite/monde'),
-                              ('Suisse & Régions',
- 'http://www.letemps.ch/rss/site/actualite/suisse_regions'),
-                              ('Sciences & Environnement',
- 'http://www.letemps.ch/rss/site/actualite/sciences_environnement'),
-                              ('Société',
- 'http://www.letemps.ch/rss/site/actualite/societe'),
-                              ('Economie & Finance',
- 'http://www.letemps.ch/rss/site/economie_finance'),
-                              ('Economie & Finance - Finance',
- 'http://www.letemps.ch/rss/site/economie_finance/finance'),
-                              ('Economie & Finance - Fonds de placement',
- 'http://www.letemps.ch/rss/site/economie_finance/fonds_placement'),
-                              ('Economie & Finance - Carrières',
- 'http://www.letemps.ch/rss/site/economie_finance/carrieres'),
-                             ('Culture',
- 'http://www.letemps.ch/rss/site/culture'),
-                              ('Culture - Cinéma',
- 'http://www.letemps.ch/rss/site/culture/cinema'),
-                              ('Culture - Musiques',
- 'http://www.letemps.ch/rss/site/culture/musiques'),
-                              ('Culture - Scènes',
- 'http://www.letemps.ch/rss/site/culture/scenes'),
-                              ('Culture - Arts plastiques',
- 'http://www.letemps.ch/rss/site/culture/arts_plastiques'),
-                              ('Livres',
- 'http://www.letemps.ch/rss/site/culture/livres'),
-                              ('Opinions',
- 'http://www.letemps.ch/rss/site/opinions'),
-                              ('Opinions - Editoriaux',
- 'http://www.letemps.ch/rss/site/opinions/editoriaux'),
-                              ('Opinions - Invités',
- 'http://www.letemps.ch/rss/site/opinions/invites'),
-                              ('Opinions - Chroniques',
- 'http://www.letemps.ch/rss/site/opinions/chroniques'),
-                              ('LifeStyle',
- 'http://www.letemps.ch/rss/site/lifestyle'),
-                              ('LifeStyle - Luxe',
- 'http://www.letemps.ch/rss/site/lifestyle/luxe'),
-                              ('LifeStyle - Horlogerie & Joaillerie',
- 'http://www.letemps.ch/rss/site/lifestyle/horlogerie_joaillerie'),
-                              ('LifeStyle - Design',
- 'http://www.letemps.ch/rss/site/lifestyle/design'),
-                              ('LifeStyle - Voyages',
- 'http://www.letemps.ch/rss/site/lifestyle/voyages'),
-                              ('LifeStyle - Gastronomie',
- 'http://www.letemps.ch/rss/site/lifestyle/gastronomie'),
-                              ('LifeStyle - Architecture & Immobilier',
- 'http://www.letemps.ch/rss/site/lifestyle/architecture_immobilier'),
-                              ('LifeStyle - Automobile',
- 'http://www.letemps.ch/rss/site/lifestyle/automobile'),
-                              ('Sports',
- 'http://www.letemps.ch/rss/site/actualite/sports'),
+                              (u'Actualit\xe9', 'http://www.letemps.ch/rss/site/'),
+                              ('Monde', 'http://www.letemps.ch/rss/site/actualite/monde'),
+                              (u'Suisse & R\xe9gions', 'http://www.letemps.ch/rss/site/actualite/suisse_regions'),
+                              ('Sciences & Environnement', 'http://www.letemps.ch/rss/site/actualite/sciences_environnement'),
+                              (u'Soci\xe9t\xe9', 'http://www.letemps.ch/rss/site/actualite/societe'),
+                              ('Economie & Finance', 'http://www.letemps.ch/rss/site/economie_finance'),
+                              ('Economie & Finance - Finance', 'http://www.letemps.ch/rss/site/economie_finance/finance'),
+                              ('Economie & Finance - Fonds de placement', 'http://www.letemps.ch/rss/site/economie_finance/fonds_placement'),
+                              (u'Economie & Finance - Carri\xe9res', 'http://www.letemps.ch/rss/site/economie_finance/carrieres'),
+                              ('Culture', 'http://www.letemps.ch/rss/site/culture'),
+                              (u'Culture - Cin\xe9ma', 'http://www.letemps.ch/rss/site/culture/cinema'),
+                              ('Culture - Musiques', 'http://www.letemps.ch/rss/site/culture/musiques'),
+                              (u'Culture - Sc\xe9nes', 'http://www.letemps.ch/rss/site/culture/scenes'),
+                              ('Culture - Arts plastiques', 'http://www.letemps.ch/rss/site/culture/arts_plastiques'),
+                              ('Livres', 'http://www.letemps.ch/rss/site/culture/livres'),
+                              ('Opinions', 'http://www.letemps.ch/rss/site/opinions'),
+                              ('Opinions - Editoriaux', 'http://www.letemps.ch/rss/site/opinions/editoriaux'),
+                              (u'Opinions - Invit\xe9s', 'http://www.letemps.ch/rss/site/opinions/invites'),
+                              ('Opinions - Chroniques', 'http://www.letemps.ch/rss/site/opinions/chroniques'),
+                              ('LifeStyle', 'http://www.letemps.ch/rss/site/lifestyle'),
+                              ('LifeStyle - Luxe', 'http://www.letemps.ch/rss/site/lifestyle/luxe'),
+                              ('LifeStyle - Horlogerie & Joaillerie', 'http://www.letemps.ch/rss/site/lifestyle/horlogerie_joaillerie'),
+                              ('LifeStyle - Design', 'http://www.letemps.ch/rss/site/lifestyle/design'),
+                              ('LifeStyle - Voyages', 'http://www.letemps.ch/rss/site/lifestyle/voyages'),
+                              ('LifeStyle - Gastronomie', 'http://www.letemps.ch/rss/site/lifestyle/gastronomie'),
+                              ('LifeStyle - Architecture & Immobilier', 'http://www.letemps.ch/rss/site/lifestyle/architecture_immobilier'),
+                              ('LifeStyle - Automobile', 'http://www.letemps.ch/rss/site/lifestyle/automobile'),
+                              ('Sports', 'http://www.letemps.ch/rss/site/actualite/sports'),
                             ]

-     def print_version(self, url):
-            return url.replace('Page', 'Facet/print')
+     def postprocess_html(self, soup, first):
+        for tag in soup.findAll('div', attrs = {'class':'box pagination'}):
+            tag.extract()
+        if not first:
+            h = soup.find('h1')
+            if h is not None:
+                h.extract()
+        return soup
+
+   #  def print_version(self, url):
+   #         return url.replace('Page', 'Facet/print')


--- a/resources/recipes/pravda.recipe
+++ b/resources/recipes/pravda.recipe
@ -0,0 +1,58 @@
+from calibre.web.feeds.news import BasicNewsRecipe
+
+class PravdaSlovakiaRecipe(BasicNewsRecipe):
+    __license__  = 'GPL v3'
+    __author__ = 'kwetal'
+    language = 'sk'
+    version = 1
+
+    title = u'Pravda'
+    publisher = u''
+    category = u'News, Newspaper'
+    description = u'News from Slovakia'
+
+    oldest_article = 1
+    max_articles_per_feed = 100
+    use_embedded_content = False
+    remove_empty_feeds = True
+
+    no_stylesheets = True
+    remove_javascript = True
+
+    # Feeds from: http://spravy.pravda.sk/info.asp?y=sk_kskop/rssinfo.htm
+    feeds = []
+    feeds.append((u'Spravodajstvo', u'http://servis.pravda.sk/rss.asp'))
+    feeds.append((u'N\u00E1zory', u'http://servis.pravda.sk/rss.asp?o=sk_nazory'))
+    feeds.append((u'\u0160port', u'http://servis.pravda.sk/rss.asp?o=sk_sport'))
+    feeds.append((u'Peniaze', u'http://servis.pravda.sk/rss.asp?o=sk_peniaze'))
+    feeds.append((u'Koktail', u'http://servis.pravda.sk/rss.asp?o=sk_koktail'))
+    feeds.append((u'Kult\u00FAra', u'http://servis.pravda.sk/rss.asp?o=sk_kultura'))
+    feeds.append((u'B\u00FDvanie', u'http://servis.pravda.sk/rss.asp?o=sk_byvanie'))
+    feeds.append((u'Veda a Technika', u'http://servis.pravda.sk/rss.asp?o=sk_veda'))
+    feeds.append((u'Mozgov\u00F1a', u'http://servis.pravda.sk/rss.asp?o=sk_mozgovna'))
+    feeds.append((u'Auto', u'http://servis.pravda.sk/rss.asp?o=sk_autoweb'))
+    feeds.append((u'Cestovanie', u'http://servis.pravda.sk/rss.asp?o=sk_cestovanie'))
+    feeds.append((u'Regi\u00F3ny', u'http://servis.pravda.sk/rss.asp?r=sk_regiony'))
+    feeds.append((u'Profesia', u'http://servis.pravda.sk/rss.asp?o=sk_profesia'))
+    feeds.append((u'Zdravie', u'http://servis.pravda.sk/rss.asp?o=sk_zdravie'))
+    feeds.append((u'\u010C\u00EDtajme de\u0165om', u'http://servis.pravda.sk/rss.asp?o=sk_citajme'))
+
+    remove_tags = []
+    remove_tags.append(dict(name = 'p', attrs = {'class': 'spatNaClanok'}))
+    remove_tags.append(dict(name = 'ul'))
+
+    extra_css = '''
+                @font-face {font-family: "serif1";src:url(res:///opt/sony/ebook/FONT/tt0011m_.ttf)}
+                @font-face {font-family: "sans1";src:url(res:///opt/sony/ebook/FONT/LiberationSans.ttf)}
+                body {font-family: sans1, serif1;}
+                .art-info {font-size: x-small; color: #696969; margin-bottom: 0.3em;}
+                .img-info {font-size: x-small; color: #696969;}
+                '''
+
+    def print_version(self, url):
+        ignore, sep, main = url.rpartition('/')
+        app, sep, id = main.rpartition('?')
+
+        app = app.replace('.asp', '')
+
+        return 'http://sport.pravda.sk/tlac.asp?r=' + app + '&' + id
--- a/resources/recipes/sme.recipe
+++ b/resources/recipes/sme.recipe
@ -0,0 +1,69 @@
+from calibre.web.feeds.news import BasicNewsRecipe
+import re
+
+class SmeRecipe(BasicNewsRecipe):
+    __license__  = 'GPL v3'
+    __author__ = 'kwetal'
+    language = 'sk'
+    version = 1
+
+    title = u'SME'
+    publisher = u''
+    category = u'News, Newspaper'
+    description = u'News from Slovakia'
+
+    oldest_article = 1
+    max_articles_per_feed = 100
+    use_embedded_content = False
+    remove_empty_feeds = True
+
+    no_stylesheets = True
+    remove_javascript = True
+
+    # Feeds from: http://rss.sme.sk/
+    feeds = []
+    feeds.append((u'Tituln\u00E1 strana', u'http://rss.sme.sk/rss/rss.asp?id=frontpage'))
+    feeds.append((u'Naj\u010D\u00EDtanej\u0161ie za 4 hodiny', u'http://rss.sme.sk/rss/rss.asp?id=smenajcit4'))
+    feeds.append((u'Naj\u010D\u00EDtanej\u0161ie za 24 hod\u00EDn', u'http://rss.sme.sk/rss/rss.asp?id=smenajcit24'))
+    feeds.append((u'Z domova', u'http://rss.sme.sk/rss/rss.asp?sek=smeonline&rub=online_zdom'))
+    feeds.append((u'Zahrani\u010Die', u'http://rss.sme.sk/rss/rss.asp?sek=smeonline&rub=online_zahr'))
+    feeds.append((u'Z domova + zahrani\u010Die', u'http://rss.sme.sk/rss/rss.asp?sek=smeonline'))
+    feeds.append((u'Ekonomika', u'http://rss.sme.sk/rss/rss.asp?sek=ekon'))
+    feeds.append((u'Kult\u00FAra', u'http://rss.sme.sk/rss/rss.asp?sek=kult'))
+    feeds.append((u'Koment\u00E1re', u'http://rss.sme.sk/rss/rss.asp?sek=koment'))
+    feeds.append((u'Volby', u'http://rss.sme.sk/rss/rss.asp?sek=eVolby'))
+    #feeds.append((u'Press foto', u'http://rss.sme.sk/rss/rss.asp?sek=smeonline&rub=online_foto'))
+    feeds.append((u'\u0160port', u'http://rss.sme.sk/rss/rss.asp?sek=sport'))
+    feeds.append((u'Futbal', u'http://rss.sme.sk/rss/rss.asp?sek=futbal'))
+    feeds.append((u'Hokej', u'http://rss.sme.sk/rss/rss.asp?sek=hokej'))
+    feeds.append((u'Po\u010D\u00EDta\u010De', u'http://rss.sme.sk/rss/rss.asp?sek=pocit'))
+    feeds.append((u'Mobil', u'http://rss.sme.sk/rss/rss.asp?sek=mobil'))
+    feeds.append((u'Veda', u'http://rss.sme.sk/rss/rss.asp?sek=veda'))
+    feeds.append((u'Natankuj', u'http://rss.sme.sk/rss/rss.asp?sek=natankuj'))
+    feeds.append((u'Auto', u'http://rss.sme.sk/rss/rss.asp?sek=auto'))
+    feeds.append((u'Dom\u00E1cnos\u0165', u'http://rss.sme.sk/rss/rss.asp?sek=domac'))
+    feeds.append((u'\u017Dena', u'http://rss.sme.sk/rss/rss.asp?sek=zena'))
+    feeds.append((u'Z\u00E1bava', u'http://rss.sme.sk/rss/rss.asp?sek=zabava'))
+    feeds.append((u'Hry', u'http://rss.sme.sk/rss/rss.asp?sek=hry'))
+    #feeds.append((u'', u''))
+
+    keep_only_tags = []
+    keep_only_tags.append(dict(name = 'div', attrs = {'id': 'contenth'}))
+    keep_only_tags.append(dict(name = 'div', attrs = {'class': 'articlec col'}))
+
+    remove_tags = []
+    remove_tags.append(dict(name = 'div', attrs = {'id': re.compile('smeplayer.*')}))
+
+    remove_tags_after = [dict(name = 'p', attrs = {'class': 'autor_line'})]
+
+    extra_css = '''
+                @font-face {font-family: "serif1";src:url(res:///opt/sony/ebook/FONT/tt0011m_.ttf)}
+                @font-face {font-family: "sans1";src:url(res:///opt/sony/ebook/FONT/LiberationSans.ttf)}
+                body {font-family: sans1, serif1;}
+                '''
+
+    def print_version(self, url):
+        parts = url.split('/')
+        id = parts[4]
+
+        return u'http://korzar.sme.sk/clanok_tlac.asp?cl=' + str(id)
--- a/resources/recipes/straitstimes.recipe
+++ b/resources/recipes/straitstimes.recipe
@ -1,4 +1,3 @@
-#!/usr/bin/env  python

 __license__   = 'GPL v3'
 __copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
@ -6,6 +5,7 @@ __copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
 www.straitstimes.com
 '''

+import re
 from calibre.web.feeds.recipes import BasicNewsRecipe

 class StraitsTimes(BasicNewsRecipe):
@ -29,9 +29,21 @@ class StraitsTimes(BasicNewsRecipe):
                            ,'publisher' : publisher
                         }

-    remove_tags = [dict(name=['object','link','map'])]
+    preprocess_regexps = [
+                           (re.compile(
+                            r'<meta name="description" content="[^"]+"\s*/?>',
+                            re.IGNORECASE|re.DOTALL),
+                            lambda m:''),
+                           (re.compile(r'<!--.+?-->', re.IGNORECASE|re.DOTALL),
+                               lambda m: ''),
+                         ]
+    remove_tags = [
+                     dict(name=['object','link','map'])
+                    ,dict(name='div',attrs={'align':'left'})
+                  ]

-    keep_only_tags = [dict(name='div', attrs={'class':['top_headline','story_text']})]
+    keep_only_tags = [dict(name='div', attrs={'class':'stleft'})]
+    remove_tags_after=dict(name='div',attrs={'class':'hr_thin'})

    feeds = [
               (u'Singapore'       , u'http://www.straitstimes.com/STI/STIFILES/rss/break_singapore.xml' )
@ -47,4 +59,3 @@ class StraitsTimes(BasicNewsRecipe):
        for item in soup.findAll(style=True):
            del item['style']
        return soup
-
--- a/resources/recipes/sueddeutsche.recipe
+++ b/resources/recipes/sueddeutsche.recipe
@ -22,18 +22,24 @@ class Sueddeutsche(BasicNewsRecipe):
    encoding = 'iso-8859-15'
    remove_javascript = True

-    keep_only_tags = [
-                        dict(name='div', attrs={'id':["artikel","contentTable"]}) ,
-                         ]
+    
    remove_tags = [ dict(name='link'), dict(name='iframe'),
-                    dict(name='div', attrs={'id':["themenbox","artikelfoot","CAD_AD","SKY_AD","NT1_AD","rechteSpalte"]}),
-                    dict(name='div', attrs={'class':["similar-article-box","artikelliste","nteaser301bg","pages closed"]}),
-                    dict(name='div', attrs={'class':["listHeader","listHeader2","hr2","item","videoBigButton"]}),
-                    dict(name='p', attrs={'class':["ressortartikeln",]}),
+                    dict(name='div', attrs={'id':["bookmarking","themenbox","artikelfoot","CAD_AD",
+						  "SKY_AD","NT1_AD","navbar1","sdesiteheader"]}),
+
+                    dict(name='div', attrs={'class':["similar-article-box","artikelliste","nteaser301bg",
+					             "pages closed","basebox right narrow"]}),
+
+                    dict(name='div', attrs={'class':["articleDistractor","listHeader","listHeader2","hr2",
+						     "item","videoBigButton","articlefooter full-column",
+                                                     "bildbanderolle full-column","footerCopy padleft5"]}),
+
+                    dict(name='p', attrs={'class':["ressortartikeln","artikelFliestext","entry-summary"]}),
                    dict(name='div', attrs={'style':["position:relative;"]}),
-                    dict(name='span', attrs={'class':["nlinkheaderteaserschwarz",]}),
-                    dict(name='table', attrs={'class':["kommentare","footer","pageBoxBot","pageAktiv","bgcontent"]}),
-                    dict(name='ul', attrs={'class':["breadcrumb","articles","activities"]}),
+                    dict(name='span', attrs={'class':["nlinkheaderteaserschwarz","artikelLink","r10000000"]}),
+                    dict(name='table', attrs={'class':["stoerBS","kommentare","footer","pageBoxBot","pageAktiv","bgcontent"]}),
+                    dict(name='ul', attrs={'class':["breadcrumb","articles","activities","sitenav"]}),
+                    dict(name='td', attrs={'class':["artikelDruckenRight"]}),
                    dict(name='p', text = "ANZEIGE")
                     ]

@ -64,8 +70,8 @@ class Sueddeutsche(BasicNewsRecipe):
             (u'Reise', u'http://suche.sueddeutsche.de/query/reise/nav/%C2%A7ressort%3AReise/sort/-docdatetime?output=rss')
             ]

-   # def print_version(self, url):
-   #     return url.replace('/text/', '/text/print.html')
+    def print_version(self, url):
+        return url.replace('/text/', '/text/print.html')



--- a/resources/recipes/wsj.recipe
+++ b/resources/recipes/wsj.recipe
@ -16,7 +16,7 @@ class WallStreetJournal(BasicNewsRecipe):
        needs_subscription = True
        language = 'en'

-        max_articles_per_feed = 10
+        max_articles_per_feed = 25
        timefmt  = ' [%a, %b %d, %Y]'
        no_stylesheets = True

--- a/setup/installer/linux/freeze.py
+++ b/setup/installer/linux/freeze.py
@ -23,13 +23,13 @@ class LinuxFreeze(Command):

        is64bit = platform.architecture()[0] == '64bit'
        arch = 'x86_64' if is64bit else 'i686'
-        ffi = '/usr/lib/libffi.so.5' if is64bit else '/usr/lib/gcc/i686-pc-linux-gnu/4.4.1/libffi.so.4'
+        ffi = '/usr/lib/gcc/x86_64-pc-linux-gnu/4.4.2/libffi.so.4' if is64bit else '/usr/lib/gcc/i686-pc-linux-gnu/4.4.1/libffi.so.4'
        stdcpp = '/usr/lib/gcc/%s-pc-linux-gnu/%s/libstdc++.so.6'%(arch, '4.4.2'
                if is64bit else '4.4.1')

        QTDIR          = '/usr/lib/qt4'
        QTDLLS         = ('QtCore', 'QtGui', 'QtNetwork', 'QtSvg', 'QtXml',
-                'QtWebKit', 'QtDBus')
+                'QtWebKit', 'QtDBus', 'QtXmlPatterns')

        binary_excludes = ['libGLcore*', 'libGL*', 'libnvidia*']

--- a/setup/installer/windows/freeze.py
+++ b/setup/installer/windows/freeze.py
@ -13,8 +13,8 @@ from setup import Command, modules, functions, basenames, __version__, \
 from setup.build_environment import msvc, MT, RC
 from setup.installer.windows.wix import WixMixIn

-QT_DIR = 'C:\\Qt\\4.5.2'
-QT_DLLS = ['Core', 'Gui', 'Network', 'Svg', 'WebKit', 'Xml', 'phonon']
+QT_DIR = 'C:\\Qt\\4.6.0'
+QT_DLLS = ['Core', 'Gui', 'Network', 'Svg', 'WebKit', 'Xml', 'XmlPatterns']
 LIBUSB_DIR       = 'C:\\libusb'
 LIBUNRAR         = 'C:\\Program Files\\UnrarDLL\\unrar.dll'
 SW               = r'C:\cygwin\home\kovid\sw'
@ -347,7 +347,6 @@ class Win32Freeze(Command, WixMixIn):
                    cmd = [msvc.cc] + xflags + ['/Tc'+src, '/Fo'+dest]
                    self.run_builder(cmd)
                exe = self.j(self.base, bname+'.exe')
-                manifest = exe+'.manifest'
                lib = dll.replace('.dll', '.lib')
                if self.newer(exe, [dest, lib, self.rc_template, __file__]):
                    self.info('Linking', bname)
--- a/src/calibre/constants.py
+++ b/src/calibre/constants.py
@ -2,7 +2,7 @@ __license__   = 'GPL v3'
 __copyright__ = '2008, Kovid Goyal kovid@kovidgoyal.net'
 __docformat__ = 'restructuredtext en'
 __appname__   = 'calibre'
-__version__   = '0.6.29'
+__version__   = '0.6.30'
 __author__    = "Kovid Goyal <kovid@kovidgoyal.net>"

 import re
--- a/src/calibre/customize/profiles.py
+++ b/src/calibre/customize/profiles.py
@ -363,7 +363,7 @@ class NookOutput(OutputProfile):
    description = _('This profile is intended for the B&N Nook.')

    # Screen size is a best guess
-    screen_size               = (600, 770)
+    screen_size               = (600, 730)
    dpi                       = 167
    fbase                     = 16
    fsizes                    = [12, 12, 14, 16, 18, 20, 22, 24]
--- a/src/calibre/customize/ui.py
+++ b/src/calibre/customize/ui.py
@ -253,7 +253,9 @@ run_plugins_on_postprocess = functools.partial(_run_filetype_plugins,

 def initialize_plugin(plugin, path_to_zip_file):
    try:
-        return plugin(path_to_zip_file)
+        p = plugin(path_to_zip_file)
+        p.initialize()
+        return p
    except Exception:
        print 'Failed to initialize plugin:', plugin.name, plugin.version
        tb = traceback.format_exc()
--- a/src/calibre/debug.py
+++ b/src/calibre/debug.py
@ -6,9 +6,9 @@ __copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
 Embedded console for debugging.
 '''

-import sys, os, pprint
+import sys, os
 from calibre.utils.config import OptionParser
-from calibre.constants import iswindows, isosx
+from calibre.constants import iswindows
 from calibre import prints

 def option_parser():
@ -61,77 +61,8 @@ def migrate(old, new):
    print 'Database migrated to', os.path.abspath(new)

 def debug_device_driver():
-    from calibre.customize.ui import device_plugins
-    from calibre.devices.scanner import DeviceScanner
-    s = DeviceScanner()
-    s.scan()
-    devices = s.devices
-    if not iswindows:
-        devices = [list(x) for x in devices]
-        for d in devices:
-            for i in range(3):
-                d[i] = hex(d[i])
-
-    print 'USB devices on system:\n', pprint.pprint(devices)
-    if iswindows:
-        wmi = __import__('wmi', globals(), locals(), [], -1)
-        drives = []
-        print 'Drives detected:'
-        print '\t', '(ID, Partitions, Drive letter)'
-        for drive in wmi.WMI(find_classes=False).Win32_DiskDrive():
-            if drive.Partitions == 0:
-                continue
-            try:
-                partition = drive.associators("Win32_DiskDriveToDiskPartition")[0]
-                logical_disk = partition.associators('Win32_LogicalDiskToPartition')[0]
-                prefix = logical_disk.DeviceID+os.sep
-                drives.append((str(drive.PNPDeviceID), drive.Index, prefix))
-            except IndexError:
-                drives.append((str(drive.PNPDeviceID), 'No mount points found'))
-        for drive in drives:
-            print '\t', drive
-    if isosx:
-        from calibre.devices.usbms.device import Device
-        raw = Device.run_ioreg()
-        open('/tmp/ioreg.txt', 'wb').write(raw)
-        print 'ioreg output saved to /tmp/ioreg.txt'
-    connected_devices = []
-    for dev in device_plugins():
-        print 'Looking for', dev.__class__.__name__
-        connected = s.is_device_connected(dev, debug=True)
-        if connected:
-            connected_devices.append(dev)
-
-    errors = {}
-    success = False
-    for dev in connected_devices:
-        print 'Device possibly connected:', dev.__class__.name
-        print 'Trying to open device...',
-        try:
-            dev.open()
-            print 'OK'
-        except:
-            import traceback
-            errors[dev] = traceback.format_exc()
-            print 'failed'
-            continue
-        success = True
-        if hasattr(dev, '_main_prefix'):
-            print 'Main memory:', repr(dev._main_prefix)
-        print 'Total space:', dev.total_space()
-        break
-    if not success and errors:
-        print 'Opening of the following devices failed'
-        for dev,msg in errors.items():
-            print dev
-            print msg
-            print
-    if isosx and os.path.exists('/tmp/ioreg.txt'):
-        print
-        print
-        print "Don't forget to send the file /tmp/ioreg.txt as well"
-        print "You can view it by typing the command: open /tmp/ioreg.txt"
-
+    from calibre.devices import debug
+    print debug()
    if iswindows:
        raw_input('Press Enter to continue...')

--- a/src/calibre/devices/init.py
+++ b/src/calibre/devices/init.py
@ -76,7 +76,9 @@ def debug():
        ioreg = None
        if isosx:
            from calibre.devices.usbms.device import Device
+            mount = repr(Device.osx_run_mount())
            ioreg = Device.run_ioreg()
+            ioreg = 'Output from mount:\n\n'+mount+'\n\n'+ioreg
        connected_devices = []
        for dev in device_plugins():
            out('Looking for', dev.__class__.__name__)
--- a/src/calibre/devices/cybookg3/driver.py
+++ b/src/calibre/devices/cybookg3/driver.py
@ -55,7 +55,6 @@ class CYBOOKG3(USBMS):

    @classmethod
    def can_handle(cls, device_info, debug=False):
-        USBMS.can_handle(device_info, debug)
        if islinux:
            return device_info[3] == 'Bookeen' and device_info[4] == 'Cybook Gen3'
        return True
@ -88,7 +87,6 @@ class CYBOOK_OPUS(CYBOOKG3):

    @classmethod
    def can_handle(cls, device_info, debug=False):
-        USBMS.can_handle(device_info, debug)
        if islinux:
            return device_info[3] == 'Bookeen'
        return True
--- a/src/calibre/devices/interface.py
+++ b/src/calibre/devices/interface.py
@ -55,7 +55,15 @@ class DevicePlugin(Plugin):
        return False

    @classmethod
-    def is_usb_connected_windows(cls, devices_on_system):
+    def print_usb_device_info(cls, info):
+        try:
+            print '\t', repr(info)
+        except:
+            import traceback
+            traceback.print_exc()
+
+    @classmethod
+    def is_usb_connected_windows(cls, devices_on_system, debug=False):

        def id_iterator():
            if hasattr(cls.VENDOR_ID, 'keys'):
@ -75,9 +83,13 @@ class DevicePlugin(Plugin):
            vid, pid = 'vid_%4.4x'%vendor_id, 'pid_%4.4x'%product_id
            vidd, pidd = 'vid_%i'%vendor_id, 'pid_%i'%product_id
            for device_id in devices_on_system:
-                if (vid in device_id or vidd in device_id) and (pid in device_id or pidd in device_id):
-                    if cls.test_bcd_windows(device_id, bcd) and cls.can_handle(device_id):
-                        return True
+                if (vid in device_id or vidd in device_id) and \
+                   (pid in device_id or pidd in device_id) and \
+                   cls.test_bcd_windows(device_id, bcd):
+                       if debug:
+                           cls.print_usb_device_info(device_id)
+                       if cls.can_handle(device_id):
+                           return True
        return False

    @classmethod
@ -97,7 +109,7 @@ class DevicePlugin(Plugin):
        :param devices_on_system: List of devices currently connected
        '''
        if iswindows:
-            return cls.is_usb_connected_windows(devices_on_system)
+            return cls.is_usb_connected_windows(devices_on_system, debug=debug)

        vendors_on_system = set([x[0] for x in devices_on_system])
        vendors = cls.VENDOR_ID if hasattr(cls.VENDOR_ID, '__len__') else [cls.VENDOR_ID]
@ -118,9 +130,11 @@ class DevicePlugin(Plugin):
                                cbcd = cls.VENDOR_ID[vid][pid]
                            else:
                                cbcd = cls.BCD
-                            if cls.test_bcd(bcd, cbcd) and cls.can_handle(dev,
-                                                            debug=debug):
-                                return True
+                            if cls.test_bcd(bcd, cbcd):
+                                if debug:
+                                    cls.print_usb_device_info(dev)
+                                if cls.can_handle(dev, debug=debug):
+                                    return True
        return False


@ -152,12 +166,6 @@ class DevicePlugin(Plugin):
        :param device_info: On windows a device ID string. On Unix a tuple of
        ``(vendor_id, product_id, bcd)``.
        '''
-        try:
-            if debug:
-                print '\t', repr(device_info)
-        except:
-            import traceback
-            traceback.print_exc()
        return True

    def open(self):
--- a/src/calibre/devices/prs505/driver.py
+++ b/src/calibre/devices/prs505/driver.py
@ -204,20 +204,20 @@ class PRS505(CLI, Device):

 class PRS700(PRS505):

-    name           = 'PRS-600/700 Device Interface'
-    description    = _('Communicate with the Sony PRS-600/700 eBook reader.')
+    name           = 'PRS-600/700/900 Device Interface'
+    description    = _('Communicate with the Sony PRS-600/700/900 eBook reader.')
    author         = 'Kovid Goyal and John Schember'
-    gui_name       = 'SONY Touch edition'
+    gui_name       = 'SONY Touch/Daily edition'
    supported_platforms = ['windows', 'osx', 'linux']

    BCD          = [0x31a]

-    WINDOWS_MAIN_MEM = re.compile('PRS-((700/)|(600&))')
-    WINDOWS_CARD_A_MEM = re.compile(r'PRS-((700/\S+:)|(600_))MS')
-    WINDOWS_CARD_B_MEM = re.compile(r'PRS-((700/\S+:)|(600_))SD')
+    WINDOWS_MAIN_MEM = re.compile('PRS-((700/)|((6|9)00&))')
+    WINDOWS_CARD_A_MEM = re.compile(r'PRS-((700/\S+:)|((6|9)00_))MS')
+    WINDOWS_CARD_B_MEM = re.compile(r'PRS-((700/\S+:)|((6|9)00_))SD')

-    OSX_MAIN_MEM   = re.compile(r'Sony PRS-((700/[^:]+)|(600)) Media')
-    OSX_CARD_A_MEM = re.compile(r'Sony PRS-((700/[^:]+:)|(600 ))MS Media')
-    OSX_CARD_B_MEM = re.compile(r'Sony PRS-((700/[^:]+:)|(600 ))SD Media')
+    OSX_MAIN_MEM   = re.compile(r'Sony PRS-((700/[^:]+)|((6|9)00)) Media')
+    OSX_CARD_A_MEM = re.compile(r'Sony PRS-((700/[^:]+:)|((6|9)00 ))MS Media')
+    OSX_CARD_B_MEM = re.compile(r'Sony PRS-((700/[^:]+:)|((6|9)00 ))SD Media')


--- a/src/calibre/devices/scanner.py
+++ b/src/calibre/devices/scanner.py
@ -40,6 +40,7 @@ class LinuxScanner(object):
            prod = os.path.join(base, 'idProduct')
            bcd = os.path.join(base, 'bcdDevice')
            man = os.path.join(base, 'manufacturer')
+            serial = os.path.join(base, 'serial')
            prod_string = os.path.join(base, 'product')
            dev = []
            try:
@ -62,6 +63,11 @@ class LinuxScanner(object):
                dev.append(open(prod_string).read().strip())
            except:
                dev.append('')
+            try:
+                dev.append(open(serial).read().strip())
+            except:
+                dev.append('')
+
            ans.add(tuple(dev))
        return ans

--- a/src/calibre/devices/usbms/device.py
+++ b/src/calibre/devices/usbms/device.py
@ -323,8 +323,16 @@ class Device(DeviceConfig, DevicePlugin):
        ioreg = '/usr/sbin/ioreg'
        if not os.access(ioreg, os.X_OK):
            ioreg = 'ioreg'
-        return subprocess.Popen((ioreg+' -w 0 -S -c IOMedia').split(),
-                                stdout=subprocess.PIPE).communicate()[0]
+        cmd = (ioreg+' -w 0 -S -c IOMedia').split()
+        for i in range(3):
+            try:
+                return subprocess.Popen(cmd,
+                                    stdout=subprocess.PIPE).communicate()[0]
+            except IOError: # Probably an interrupted system call
+                if i == 2:
+                    raise
+            time.sleep(2)
+

    def osx_sort_names(self, names):
        return names
@ -372,14 +380,28 @@ class Device(DeviceConfig, DevicePlugin):
                break
        return self.osx_sort_names(names)

+    @classmethod
+    def osx_run_mount(cls):
+        for i in range(3):
+            try:
+                return subprocess.Popen('mount',
+                                    stdout=subprocess.PIPE).communicate()[0]
+            except IOError: # Probably an interrupted system call
+                if i == 2:
+                    raise
+            time.sleep(2)
+
    def open_osx(self):
-        mount = subprocess.Popen('mount', shell=True,  stdout=subprocess.PIPE).stdout.read()
+        mount = self.osx_run_mount()
        names = self.get_osx_mountpoints()
        dev_pat = r'/dev/%s(\w*)\s+on\s+([^\(]+)\s+'
        if 'main' not in names.keys():
            raise DeviceError(_('Unable to detect the %s disk drive. Try rebooting.')%self.__class__.__name__)
        main_pat = dev_pat % names['main']
-        self._main_prefix = re.search(main_pat, mount).group(2) + os.sep
+        main_match = re.search(main_pat, mount)
+        if main_match is None:
+            raise DeviceError(_('Unable to detect the %s mount point. Try rebooting.')%self.__class__.__name__)
+        self._main_prefix = main_match.group(2) + os.sep
        card_a_pat = names['carda'] if 'carda' in names.keys() else None
        card_b_pat = names['cardb'] if 'cardb' in names.keys() else None

--- a/src/calibre/devices/usbobserver/Makefile
+++ b/src/calibre/devices/usbobserver/Makefile
@ -2,11 +2,7 @@ usbobserver.so : usbobserver.o
 	gcc -arch i386 -arch ppc -bundle usbobserver.o -o usbobserver.so -framework Python -framework IOKit -framework CoreFoundation

 usbobserver.o : usbobserver.c
-	gcc -arch i386 -arch ppc -dynamic -I/Library/Frameworks/Python.framework/Versions/2.5/include/python2.5 -c usbobserver.c -o usbobserver.o 
-
-
-install : usbobserver.so
-	cp usbobserver.so /Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/
+	gcc -arch i386 -arch ppc -dynamic -I/Library/Frameworks/Python.framework/Versions/Current/Headers -c usbobserver.c -o usbobserver.o 

 clean : 
 	rm -f *.o *.so
--- a/src/calibre/devices/usbobserver/usbobserver.c
+++ b/src/calibre/devices/usbobserver/usbobserver.c
@ -26,28 +26,34 @@

 #include <IOKit/usb/IOUSBLib.h>
 #include <IOKit/IOCFPlugIn.h>
+#include <IOKit/IOKitLib.h>
 #include <mach/mach.h>

+CFStringRef USB_PROPS[3] = { CFSTR("USB Vendor Name"), CFSTR("USB Product Name"), CFSTR("USB Serial Number") };
+
+static PyObject*
+get_iokit_string_property(io_service_t dev, int prop) {
+  CFTypeRef PropRef;
+  char buf[500];
+
+  PropRef = IORegistryEntryCreateCFProperty(dev, USB_PROPS[prop], kCFAllocatorDefault, 0);
+  if (PropRef) {
+      if(!CFStringGetCString(PropRef, buf, 500, kCFStringEncodingUTF8)) buf[0] = '\0';
+  } else buf[0] = '\0';
+
+  return PyUnicode_DecodeUTF8(buf, strlen(buf), "replace");
+}
+
 static PyObject *
 usbobserver_get_usb_devices(PyObject *self, PyObject *args) {
  
-  mach_port_t masterPort;
  CFMutableDictionaryRef matchingDict;
  kern_return_t kr;

-  /* Create a master port for communication with IOKit */
-  kr = IOMasterPort(MACH_PORT_NULL, &masterPort);
-
-  if (kr || !masterPort) {
-    PyErr_SetString(PyExc_RuntimeError, "Couldn't create master IOKit port");
-    return NULL;
-  }
-  
  //Set up matching dictionary for class IOUSBDevice and its subclasses
  matchingDict = IOServiceMatching(kIOUSBDeviceClassName);
  if (!matchingDict) {
    PyErr_SetString(PyExc_RuntimeError, "Couldn't create a USB matching dictionary");
-    mach_port_deallocate(mach_task_self(), masterPort);
    return NULL;
  }

@ -58,12 +64,12 @@ usbobserver_get_usb_devices(PyObject *self, PyObject *args) {
  SInt32 score;
  IOUSBDeviceInterface182 **dev = NULL;
  UInt16 vendor, product, bcd;
+  PyObject *manufacturer, *productn, *serial;

  PyObject *devices, *device;
  devices = PyList_New(0);
  if (devices == NULL) {
      PyErr_NoMemory();
-      mach_port_deallocate(mach_task_self(), masterPort);
      return NULL;
  }

@ -85,7 +91,15 @@ usbobserver_get_usb_devices(PyObject *self, PyObject *args) {
    kr = (*dev)->GetDeviceVendor(dev, &vendor);
    kr = (*dev)->GetDeviceProduct(dev, &product);
    kr = (*dev)->GetDeviceReleaseNumber(dev, &bcd);
-    device = Py_BuildValue("(iii)", vendor, product, bcd);
+
+    manufacturer = get_iokit_string_property(usbDevice, 0);
+    if (manufacturer == NULL) manufacturer = Py_None;
+    productn = get_iokit_string_property(usbDevice, 1);
+    if (productn == NULL) productn = Py_None;
+    serial = get_iokit_string_property(usbDevice, 2);
+    if (serial == NULL) serial = Py_None;
+
+    device = Py_BuildValue("(iiiNNN)", vendor, product, bcd, manufacturer, productn, serial);
    if (device == NULL) {
      IOObjectRelease(usbDevice);
      (*plugInInterface)->Release(plugInInterface);
@ -109,11 +123,7 @@ usbobserver_get_usb_devices(PyObject *self, PyObject *args) {
    Py_DECREF(device);
  }
    
-
-  //Finished with master port
-  mach_port_deallocate(mach_task_self(), masterPort);
-  
-  return Py_BuildValue("N", devices);
+  return devices;
 }

 static PyMethodDef usbobserver_methods[] = {
--- a/src/calibre/ebooks/lrf/output.py
+++ b/src/calibre/ebooks/lrf/output.py
@ -33,7 +33,7 @@ class LRFOptions(object):
            if unicode(x.file_as):
                self.title_sort = unicode(x.file_as)
        self.freetext = f2s(m.description)
-        self.category = f2s(m.tags)
+        self.category = f2s(m.subject)
        self.cover = None
        self.use_metadata_cover = True
        self.output = output
--- a/src/calibre/ebooks/metadata/init.py
+++ b/src/calibre/ebooks/metadata/init.py
@ -386,3 +386,36 @@ class MetaInformation(object):

    def __nonzero__(self):
        return bool(self.title or self.author or self.comments or self.tags)
+
+def check_isbn10(isbn):
+    try:
+        digits = map(int, isbn[:9])
+        products = [(i+1)*digits[i] for i in range(9)]
+        check = sum(products)%11
+        if (check == 10 and isbn[9] == 'X') or check == int(isbn[9]):
+            return isbn
+    except:
+        pass
+    return None
+
+def check_isbn13(isbn):
+    try:
+        digits = map(int, isbn[:12])
+        products = [(1 if i%2 ==0 else 3)*digits[i] for i in range(12)]
+        check = 10 - (sum(products)%10)
+        if check == 10:
+            check = 0
+        if str(check) == isbn[12]:
+            return isbn
+    except:
+        pass
+    return None
+
+def check_isbn(isbn):
+    isbn = re.sub(r'[^0-9X]', '', isbn.upper())
+    if len(isbn) == 10:
+        return check_isbn10(isbn)
+    if len(isbn) == 13:
+        return check_isbn13(isbn)
+    return None
+
--- a/src/calibre/ebooks/metadata/meta.py
+++ b/src/calibre/ebooks/metadata/meta.py
@ -121,6 +121,7 @@ def metadata_from_filename(name, pat=None):
    mi = MetaInformation(None, None)
    if pat is None:
        pat = re.compile(prefs.get('filename_pattern'))
+    name = name.replace('_', ' ')
    match = pat.search(name)
    if match:
        try:
@ -131,6 +132,15 @@ def metadata_from_filename(name, pat=None):
            au = match.group('author')
            aus = string_to_authors(au)
            mi.authors = aus
+            if prefs['swap_author_names'] and mi.authors:
+                def swap(a):
+                    parts = a.split()
+                    if len(parts) > 1:
+                        t = parts[-1]
+                        parts = parts[:-1]
+                        parts.insert(0, t)
+                    return ' '.join(parts)
+                mi.authors = [swap(x) for x in mi.authors]
        except (IndexError, ValueError):
            pass
        try:
--- a/src/calibre/ebooks/mobi/reader.py
+++ b/src/calibre/ebooks/mobi/reader.py
@ -435,7 +435,8 @@ class MobiReader(object):
                open(os.path.splitext(htmlfile)[0] + '.ncx', 'wb').write(ncx)

    def read_embedded_metadata(self, root, elem, guide):
-        raw = '<package>' + html.tostring(elem, encoding='utf-8') + '</package>'
+        raw = '<?xml version="1.0" encoding="utf-8" ?>\n<package>' + \
+                html.tostring(elem, encoding='utf-8') + '</package>'
        stream = cStringIO.StringIO(raw)
        opf = OPF(stream)
        self.embedded_mi = MetaInformation(opf)
@ -602,7 +603,7 @@ class MobiReader(object):
                * opf.cover.split('/'))):
                opf.cover = None

-        manifest = [(htmlfile, 'text/x-oeb1-document'),
+        manifest = [(htmlfile, 'application/xhtml+xml'),
            (os.path.abspath('styles.css'), 'text/css')]
        bp = os.path.dirname(htmlfile)
        for i in getattr(self, 'image_names', []):
--- a/src/calibre/gui2/init.py
+++ b/src/calibre/gui2/init.py
@ -385,6 +385,7 @@ def initialize_file_icon_provider():

 def file_icon_provider():
    global _file_icon_provider
+    initialize_file_icon_provider()
    return _file_icon_provider

 class FileDialog(QObject):
--- a/src/calibre/gui2/add.py
+++ b/src/calibre/gui2/add.py
@ -24,6 +24,7 @@ class DuplicatesAdder(QThread):
    def run(self):
        count = 1
        for mi, cover, formats in self.duplicates:
+            formats = [f for f in formats if not f.lower().endswith('.opf')]
            id = self.db.create_book_entry(mi, cover=cover,
                    add_duplicates=True)
            self.db_adder.add_formats(id, formats)
@ -139,6 +140,7 @@ class DBAdder(Thread):
            if id is None:
                self.duplicates.append((mi, cover, formats))
            else:
+                formats = [f for f in formats if not f.lower().endswith('.opf')]
                self.add_formats(id, formats)
        else:
            self.names.append(name)
--- a/src/calibre/gui2/dialogs/config/add_save.py
+++ b/src/calibre/gui2/dialogs/config/add_save.py
@ -56,6 +56,7 @@ class AddSave(QTabWidget, Ui_TabWidget):
        self.opt_read_metadata_from_filename.setChecked(not prefs['read_file_metadata'])
        self.filename_pattern = FilenamePattern(self)
        self.metadata_box.layout().insertWidget(0, self.filename_pattern)
+        self.opt_swap_author_names.setChecked(prefs['swap_author_names'])

    def validate(self):
        tmpl = preprocess_template(self.opt_template.text())
@ -87,6 +88,7 @@ class AddSave(QTabWidget, Ui_TabWidget):
        prefs['read_file_metadata'] = not bool(self.opt_read_metadata_from_filename.isChecked())
        pattern = self.filename_pattern.commit()
        prefs['filename_pattern'] = pattern
+        prefs['swap_author_names'] = bool(self.opt_swap_author_names.isChecked())

        return True

--- a/src/calibre/gui2/dialogs/config/add_save.ui
+++ b/src/calibre/gui2/dialogs/config/add_save.ui
@ -20,8 +20,8 @@
   <attribute name="title">
    <string>&amp;Adding books</string>
   </attribute>
-   <layout class="QVBoxLayout" name="verticalLayout_2">
-    <item>
+   <layout class="QGridLayout" name="gridLayout_3">
+    <item row="0" column="0" colspan="2">
     <widget class="QLabel" name="label_6">
      <property name="text">
       <string>Here you can control how calibre will read metadata from the files you add to it. calibre can either read metadata from the contents of the file, or from the filename.</string>
@ -31,14 +31,24 @@
      </property>
     </widget>
    </item>
-    <item>
+    <item row="1" column="0">
     <widget class="QCheckBox" name="opt_read_metadata_from_filename">
      <property name="text">
       <string>Read metadata only from &amp;file name</string>
      </property>
     </widget>
    </item>
-    <item>
+    <item row="1" column="1">
+     <widget class="QCheckBox" name="opt_swap_author_names">
+      <property name="toolTip">
+       <string>Swap the firstname and lastname of the author. This affects only metadata read from file names.</string>
+      </property>
+      <property name="text">
+       <string>&amp;Swap author firstname and lastname</string>
+      </property>
+     </widget>
+    </item>
+    <item row="2" column="0" colspan="2">
     <widget class="QGroupBox" name="metadata_box">
      <property name="title">
       <string>&amp;Configure metadata from file name</string>
--- a/src/calibre/gui2/dialogs/metadata_single.py
+++ b/src/calibre/gui2/dialogs/metadata_single.py
@ -23,7 +23,8 @@ from calibre.gui2.dialogs.fetch_metadata import FetchMetadata
 from calibre.gui2.dialogs.tag_editor import TagEditor
 from calibre.gui2.widgets import ProgressIndicator
 from calibre.ebooks import BOOK_EXTENSIONS
-from calibre.ebooks.metadata import authors_to_sort_string, string_to_authors, authors_to_string
+from calibre.ebooks.metadata import authors_to_sort_string, string_to_authors, \
+        authors_to_string, check_isbn
 from calibre.ebooks.metadata.library_thing import cover_from_isbn
 from calibre import islinux
 from calibre.ebooks.metadata.meta import get_metadata
@ -336,6 +337,7 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog):
        isbn = db.isbn(self.id, index_is_id=True)
        if not isbn:
            isbn = ''
+        self.isbn.textChanged.connect(self.validate_isbn)
        self.isbn.setText(isbn)
        aus = self.db.author_sort(row)
        self.author_sort.setText(aus if aus else '')
@ -380,6 +382,7 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog):
            self.series_index.setValue(1.0)
        QObject.connect(self.series, SIGNAL('currentIndexChanged(int)'), self.enable_series_index)
        QObject.connect(self.series, SIGNAL('editTextChanged(QString)'), self.enable_series_index)
+        self.series.lineEdit().editingFinished.connect(self.increment_series_index)

        self.show()
        height_of_rest = self.frameGeometry().height() - self.cover.height()
@ -394,6 +397,20 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog):
                self.cover.setPixmap(pm)
            self.cover_data = cover

+    def validate_isbn(self, isbn):
+        isbn = unicode(isbn).strip()
+        if not isbn:
+            self.isbn.setStyleSheet('QLineEdit { background-color: rgba(0,255,0,0%) }')
+            self.isbn.setToolTip(_('This ISBN number is valid'))
+            return
+
+        if check_isbn(isbn):
+            self.isbn.setStyleSheet('QLineEdit { background-color: rgba(0,255,0,20%) }')
+            self.isbn.setToolTip(_('This ISBN number is valid'))
+        else:
+            self.isbn.setStyleSheet('QLineEdit { background-color: rgba(255,0,0,20%) }')
+            self.isbn.setToolTip(_('This ISBN number is invalid'))
+
    def show_format(self, item, *args):
        fmt = item.ext
        self.emit(SIGNAL('view_format(PyQt_PyObject)'), fmt)
@ -581,6 +598,16 @@ class MetadataSingleDialog(ResizableDialog, Ui_MetadataSingleDialog):
    def enable_series_index(self, *args):
        self.series_index.setEnabled(True)

+    def increment_series_index(self):
+        if self.db is not None:
+            try:
+                series = unicode(self.series.text())
+                if series:
+                    ns = self.db.get_next_series_num_for(series)
+                    self.series_index.setValue(ns)
+            except:
+                traceback.print_exc()
+
    def remove_unused_series(self):
        self.db.remove_unused_series()
        idx = qstring_to_unicode(self.series.currentText())
--- a/src/calibre/gui2/dialogs/select_formats.py
+++ b/src/calibre/gui2/dialogs/select_formats.py
@ -0,0 +1,77 @@
+#!/usr/bin/env python
+# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:ai
+from __future__ import with_statement
+
+__license__   = 'GPL v3'
+__copyright__ = '2009, Kovid Goyal <kovid@kovidgoyal.net>'
+__docformat__ = 'restructuredtext en'
+
+
+from PyQt4.Qt import QVBoxLayout, QDialog, QLabel, QDialogButtonBox, Qt, \
+        QAbstractListModel, QVariant, QListView, QSize
+
+from calibre.gui2 import NONE, file_icon_provider
+
+class Formats(QAbstractListModel):
+
+    def __init__(self, fmts):
+        QAbstractListModel.__init__(self)
+        self.fmts = sorted(fmts)
+        self.fi = file_icon_provider()
+
+    def rowCount(self, parent):
+        return len(self.fmts)
+
+    def data(self, index, role):
+        row = index.row()
+        if role == Qt.DisplayRole:
+            return QVariant(self.fmts[row].upper())
+        if role == Qt.DecorationRole:
+            return QVariant(self.fi.icon_from_ext(self.fmts[row].lower()))
+        return NONE
+
+    def flags(self, index):
+        return Qt.ItemIsSelectable|Qt.ItemIsEnabled
+
+    def fmt(self, idx):
+        return self.fmts[idx.row()]
+
+class SelectFormats(QDialog):
+
+    def __init__(self, fmt_list, msg, single=False, parent=None):
+        QDialog.__init__(self, parent)
+        self._l = QVBoxLayout(self)
+        self.setLayout(self._l)
+        self.setWindowTitle(_('Choose formats'))
+        self._m = QLabel(msg)
+        self._m.setWordWrap = True
+        self._l.addWidget(self._m)
+        self.formats = Formats(fmt_list)
+        self.fview = QListView(self)
+        self._l.addWidget(self.fview)
+        self.fview.setModel(self.formats)
+        self.fview.setSelectionMode(self.fview.SingleSelection if single else
+                self.fview.MultiSelection)
+        self.bbox = \
+        QDialogButtonBox(QDialogButtonBox.Ok|QDialogButtonBox.Cancel,
+                Qt.Horizontal, self)
+        self._l.addWidget(self.bbox)
+        self.bbox.accepted.connect(self.accept)
+        self.bbox.rejected.connect(self.reject)
+        self.fview.setIconSize(QSize(48, 48))
+        self.fview.setSpacing(2)
+
+        self.resize(350, 500)
+        self.selected_formats = set([])
+
+    def accept(self, *args):
+        for idx in self.fview.selectedIndexes():
+            self.selected_formats.add(self.formats.fmt(idx))
+        QDialog.accept(self, *args)
+
+if __name__ == '__main__':
+    from PyQt4.Qt import QApplication
+    app = QApplication([])
+    d = SelectFormats(['epub', 'lrf', 'lit', 'mobi'], 'Choose a format')
+    d.exec_()
+    print d.selected_formats
--- a/src/calibre/gui2/library.py
+++ b/src/calibre/gui2/library.py
@ -11,7 +11,7 @@ from PyQt4.QtGui import QTableView, QAbstractItemView, QColor, \
                        QPen, QStyle, QPainter, \
                        QImage, QApplication, QMenu, \
                        QStyledItemDelegate, QCompleter
-from PyQt4.QtCore import QAbstractTableModel, QVariant, Qt, \
+from PyQt4.QtCore import QAbstractTableModel, QVariant, Qt, pyqtSignal, \
                         SIGNAL, QObject, QSize, QModelIndex, QDate

 from calibre import strftime
@ -155,6 +155,10 @@ class TagsDelegate(QStyledItemDelegate):
        return editor

 class BooksModel(QAbstractTableModel):
+
+    about_to_be_sorted = pyqtSignal(object, name='aboutToBeSorted')
+    sorting_done       = pyqtSignal(object, name='sortingDone')
+
    headers = {
                        'title'     : _("Title"),
                        'authors'   : _("Author(s)"),
@ -285,13 +289,14 @@ class BooksModel(QAbstractTableModel):
    def sort(self, col, order, reset=True):
        if not self.db:
            return
+        self.about_to_be_sorted.emit(self.db.id)
        ascending = order == Qt.AscendingOrder
        self.db.sort(self.column_map[col], ascending)
        if reset:
            self.clear_caches()
            self.reset()
        self.sorted_on = (self.column_map[col], order)
-
+        self.sorting_done.emit(self.db.index)

    def refresh(self, reset=True):
        try:
@ -631,12 +636,16 @@ class BooksModel(QAbstractTableModel):
                val *= 2
                self.db.set_rating(id, val)
            elif column == 'series':
+                val = val.strip()
                pat = re.compile(r'\[([.0-9]+)\]')
                match = pat.search(val)
                if match is not None:
                    self.db.set_series_index(id, float(match.group(1)))
                    val = pat.sub('', val)
-                val = val.strip()
+                elif val:
+                    ni = self.db.get_next_series_num_for(val)
+                    if ni != 1:
+                        self.db.set_series_index(id, ni)
                if val:
                    self.db.set_series(id, val)
            elif column == 'timestamp':
@ -696,6 +705,22 @@ class BooksView(TableView):
        hv = self.verticalHeader()
        hv.setClickable(True)
        hv.setCursor(Qt.PointingHandCursor)
+        self.selected_ids = []
+        self._model.about_to_be_sorted.connect(self.about_to_be_sorted)
+        self._model.sorting_done.connect(self.sorting_done)
+
+    def about_to_be_sorted(self, idc):
+        selected_rows = [r.row() for r in self.selectionModel().selectedRows()]
+        self.selected_ids = [idc(r) for r in selected_rows]
+
+    def sorting_done(self, indexc):
+        if self.selected_ids:
+            indices = [self.model().index(indexc(i), 0) for i in
+                    self.selected_ids]
+            sm = self.selectionModel()
+            for idx in indices:
+                sm.select(idx, sm.Select|sm.Rows)
+        self.selected_ids = []

    def columns_sorted(self):
        for i in range(self.model().columnCount(None)):
--- a/src/calibre/gui2/ui.py
+++ b/src/calibre/gui2/ui.py
@ -314,6 +314,16 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
        self.view_menu.addAction(_('View'))
        self.view_menu.addAction(_('View specific format'))
        self.action_view.setMenu(self.view_menu)
+
+        self.delete_menu = QMenu()
+        self.delete_menu.addAction(_('Remove selected books'))
+        self.delete_menu.addAction(
+                _('Remove files of a specific format from selected books..'))
+        self.delete_menu.addAction(
+                _('Remove all formats from selected books, except...'))
+        self.delete_menu.addAction(
+                _('Remove covers from selected books'))
+        self.action_del.setMenu(self.delete_menu)
        QObject.connect(self.action_save, SIGNAL("triggered(bool)"),
                self.save_to_disk)
        QObject.connect(self.save_menu.actions()[0], SIGNAL("triggered(bool)"),
@ -330,6 +340,11 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
                SIGNAL("triggered(bool)"), self.view_specific_format)
        self.connect(self.action_open_containing_folder,
                SIGNAL('triggered(bool)'), self.view_folder)
+        self.delete_menu.actions()[0].triggered.connect(self.delete_books)
+        self.delete_menu.actions()[1].triggered.connect(self.delete_selected_formats)
+        self.delete_menu.actions()[2].triggered.connect(self.delete_all_but_selected_formats)
+        self.delete_menu.actions()[3].triggered.connect(self.delete_covers)
+
        self.action_open_containing_folder.setShortcut(Qt.Key_O)
        self.addAction(self.action_open_containing_folder)
        self.action_sync.setShortcut(Qt.Key_D)
@ -376,6 +391,8 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
                setPopupMode(QToolButton.MenuButtonPopup)
        self.tool_bar.widgetForAction(self.action_view).\
                setPopupMode(QToolButton.MenuButtonPopup)
+        self.tool_bar.widgetForAction(self.action_del).\
+                setPopupMode(QToolButton.MenuButtonPopup)
        self.tool_bar.widgetForAction(self.action_preferences).\
                setPopupMode(QToolButton.MenuButtonPopup)
        self.tool_bar.setContextMenuPolicy(Qt.PreventContextMenu)
@ -987,7 +1004,72 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
    ############################################################################

    ############################### Delete books ###############################
-    def delete_books(self, checked):
+
+    def _get_selected_formats(self, msg):
+        from calibre.gui2.dialogs.select_formats import SelectFormats
+        fmts = self.library_view.model().db.all_formats()
+        d = SelectFormats([x.lower() for x in fmts], msg, parent=self)
+        if d.exec_() != d.Accepted:
+            return None
+        return d.selected_formats
+
+    def _get_selected_ids(self, err_title=_('Cannot delete')):
+        rows = self.library_view.selectionModel().selectedRows()
+        if not rows or len(rows) == 0:
+            d = error_dialog(self, err_title, _('No book selected'))
+            d.exec_()
+            return set([])
+        return set(map(self.library_view.model().id, rows))
+
+    def delete_selected_formats(self, *args):
+        ids = self._get_selected_ids()
+        if not ids:
+            return
+        fmts = self._get_selected_formats(
+            _('Choose formats to be deleted'))
+        if not fmts:
+            return
+        for id in ids:
+            for fmt in fmts:
+                self.library_view.model().db.remove_format(id, fmt,
+                        index_is_id=True, notify=False)
+        self.library_view.model().refresh_ids(ids)
+        self.library_view.model().current_changed(self.library_view.currentIndex(),
+                self.library_view.currentIndex())
+
+    def delete_all_but_selected_formats(self, *args):
+        ids = self._get_selected_ids()
+        if not ids:
+            return
+        fmts = self._get_selected_formats(
+            '<p>'+_('Choose formats <b>not</b> to be deleted'))
+        if fmts is None:
+            return
+        for id in ids:
+            bfmts = self.library_view.model().db.formats(id, index_is_id=True)
+            if bfmts is None:
+                continue
+            bfmts = set([x.lower() for x in bfmts.split(',')])
+            rfmts = bfmts - set(fmts)
+            for fmt in rfmts:
+                self.library_view.model().db.remove_format(id, fmt,
+                        index_is_id=True, notify=False)
+        self.library_view.model().refresh_ids(ids)
+        self.library_view.model().current_changed(self.library_view.currentIndex(),
+                self.library_view.currentIndex())
+
+
+    def delete_covers(self, *args):
+        ids = self._get_selected_ids()
+        if not ids:
+            return
+        for id in ids:
+            self.library_view.model().db.remove_cover(id)
+        self.library_view.model().refresh_ids(ids)
+        self.library_view.model().current_changed(self.library_view.currentIndex(),
+                self.library_view.currentIndex())
+
+    def delete_books(self, *args):
        '''
        Delete selected books from device or library.
        '''
@ -1591,7 +1673,6 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
                    view.resizeColumnsToContents()
                view.resize_on_select = False
        self.status_bar.reset_info()
-        self.current_view().clearSelection()
        if location == 'library':
            self.action_edit.setEnabled(True)
            self.action_convert.setEnabled(True)
@ -1600,6 +1681,8 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
            self.action_sync.setEnabled(True)
            self.status_bar.tag_view_button.setEnabled(True)
            self.status_bar.cover_flow_button.setEnabled(True)
+            for action in list(self.delete_menu.actions())[1:]:
+                action.setEnabled(True)
        else:
            self.action_edit.setEnabled(False)
            self.action_convert.setEnabled(False)
@ -1608,6 +1691,8 @@ class Main(MainWindow, Ui_MainWindow, DeviceGUI):
            self.action_sync.setEnabled(False)
            self.status_bar.tag_view_button.setEnabled(False)
            self.status_bar.cover_flow_button.setEnabled(False)
+            for action in list(self.delete_menu.actions())[1:]:
+                action.setEnabled(False)


    def device_job_exception(self, job):
--- a/src/calibre/gui2/viewer/documentview.py
+++ b/src/calibre/gui2/viewer/documentview.py
@ -568,8 +568,8 @@ class DocumentView(QWebView):
        delta_y = self.document.window_height - 25
        if self.document.at_top:
            if self.manager is not None:
-                self.manager.previous_document()
                self.to_bottom = True
+                self.manager.previous_document()
        else:
            opos = self.document.ypos
            upper_limit = opos - delta_y
--- a/src/calibre/library/database2.py
+++ b/src/calibre/library/database2.py
@ -10,6 +10,7 @@ import os, re, sys, shutil, cStringIO, glob, collections, textwrap, \
       itertools, functools, traceback
 from itertools import repeat
 from datetime import datetime
+from math import floor

 from PyQt4.QtCore import QThread, QReadWriteLock
 try:
@ -864,6 +865,11 @@ class LibraryDatabase2(LibraryDatabase):
        path = os.path.join(self.library_path, self.path(id, index_is_id=True), 'cover.jpg')
        return os.access(path, os.R_OK)

+    def remove_cover(self, id):
+        path = os.path.join(self.library_path, self.path(id, index_is_id=True), 'cover.jpg')
+        if os.path.exists(path):
+            os.remove(path)
+
    def set_cover(self, id, data):
        '''
        Set the cover for this book.
@ -1080,7 +1086,18 @@ class LibraryDatabase2(LibraryDatabase):
                    if tags and tag in tags.lower():
                        yield r[FIELD_MAP['id']]

-
+    def get_next_series_num_for(self, series):
+        series_id = self.conn.get('SELECT id from series WHERE name=?',
+                (series,), all=False)
+        if series_id is None:
+            return 1.0
+        series_num = self.conn.get(
+            ('SELECT MAX(series_index) FROM books WHERE id IN '
+            '(SELECT book FROM books_series_link where series=?)'),
+            (series_id,), all=False)
+        if series_num is None:
+            return 1.0
+        return floor(series_num+1)

    def set(self, row, column, val):
        '''
--- a/src/calibre/library/sqlite.py
+++ b/src/calibre/library/sqlite.py
@ -8,7 +8,7 @@ Wrapper for multi-threaded access to a single sqlite database connection. Serial
 all calls.
 '''
 import sqlite3 as sqlite, traceback, time, uuid
-from sqlite3 import IntegrityError
+from sqlite3 import IntegrityError, OperationalError
 from threading import Thread
 from Queue import Queue
 from threading import RLock
@ -139,7 +139,16 @@ class DBThread(Thread):
                else:
                    func = getattr(self.conn, func)
                    try:
-                        ok, res = True, func(*args, **kwargs)
+                        for i in range(3):
+                            try:
+                                ok, res = True, func(*args, **kwargs)
+                                break
+                            except OperationalError, err:
+                                # Retry if unable to open db file
+                                if 'unable to open' not in str(err) or i == 2:
+                                    raise
+                                traceback.print_exc()
+                            time.sleep(0.5)
                    except Exception, err:
                        ok, res = False, (err, traceback.format_exc())
                self.results.put((ok, res))
--- a/src/calibre/translations/ar.po
+++ b/src/calibre/translations/ar.po
--- a/src/calibre/translations/ca.po
+++ b/src/calibre/translations/ca.po
--- a/src/calibre/translations/calibre.pot
+++ b/src/calibre/translations/calibre.pot
--- a/src/calibre/translations/de.po
+++ b/src/calibre/translations/de.po
--- a/src/calibre/translations/es.po
+++ b/src/calibre/translations/es.po
--- a/src/calibre/translations/fr.po
+++ b/src/calibre/translations/fr.po
--- a/src/calibre/translations/hr.po
+++ b/src/calibre/translations/hr.po
--- a/src/calibre/translations/hu.po
+++ b/src/calibre/translations/hu.po
--- a/src/calibre/translations/it.po
+++ b/src/calibre/translations/it.po
--- a/src/calibre/translations/lv.po
+++ b/src/calibre/translations/lv.po
--- a/src/calibre/translations/nb.po
+++ b/src/calibre/translations/nb.po
--- a/src/calibre/translations/nl.po
+++ b/src/calibre/translations/nl.po
--- a/src/calibre/translations/pt_BR.po
+++ b/src/calibre/translations/pt_BR.po
--- a/src/calibre/translations/ru.po
+++ b/src/calibre/translations/ru.po
--- a/src/calibre/translations/uk.po
+++ b/src/calibre/translations/uk.po
--- a/src/calibre/translations/zh_TW.po
+++ b/src/calibre/translations/zh_TW.po
--- a/src/calibre/utils/config.py
+++ b/src/calibre/utils/config.py
@ -649,6 +649,8 @@ def _prefs():
              help=_('Read metadata from files'))
    c.add_opt('worker_process_priority', default='normal',
              help=_('The priority of worker processes'))
+    c.add_opt('swap_author_names', default=False,
+            help=_('Swap author first and last names when reading metadata'))

    c.add_opt('migrated', default=False, help='For Internal use. Don\'t modify.')
    return c
--- a/src/calibre/utils/localization.py
+++ b/src/calibre/utils/localization.py
@ -100,6 +100,8 @@ _extra_lang_codes = {
        'en_CA' : _('English (CA)'),
        'en_IN' : _('English (IND)'),
        'en_TH' : _('English (TH)'),
+        'en_CY' : _('English (CY)'),
+        'de_AT' : _('German (AT)'),
        'nl'    : _('Dutch (NL)'),
        'nl_BE' : _('Dutch (BE)'),
        'und'   : _('Unknown')
--- a/src/calibre/web/feeds/input.py
+++ b/src/calibre/web/feeds/input.py
@ -38,6 +38,9 @@ class RecipeInput(InputFormatPlugin):
        OptionRecommendation(name='password', recommended_value=None,
            help=_('Password for sites that require a login to access '
                'content.')),
+        OptionRecommendation(name='dont_download_recipe',
+            recommended_value=False,
+            help=_('Download latest version of builtin recipes')),
        OptionRecommendation(name='lrf', recommended_value=False,
            help='Optimize fetching for subsequent conversion to LRF.'),
        ])
@ -52,7 +55,8 @@ class RecipeInput(InputFormatPlugin):
        else:
            title = getattr(opts, 'original_recipe_input_arg', recipe_or_file)
            title = os.path.basename(title).rpartition('.')[0]
-            raw = get_builtin_recipe_by_title(title, log=log, download_recipe=True)
+            raw = get_builtin_recipe_by_title(title, log=log,
+                    download_recipe=not opts.dont_download_recipe)
            builtin = False
            try:
                recipe = compile_recipe(raw)
@ -68,6 +72,8 @@ class RecipeInput(InputFormatPlugin):
            if builtin:
                raw = get_builtin_recipe_by_title(title, log=log,
                        download_recipe=False)
+                if raw is None:
+                    raise ValueError('Failed to find builtin recipe: '+title)
                recipe = compile_recipe(raw)


--- a/src/calibre/web/feeds/recipes/collection.py
+++ b/src/calibre/web/feeds/recipes/collection.py
@ -20,7 +20,7 @@ NS = 'http://calibre-ebook.com/recipe_collection'
 E = ElementMaker(namespace=NS, nsmap={None:NS})

 def iterate_over_builtin_recipe_files():
-    exclude = ['craigslist', 'iht', 'le_temps', 'outlook_india', 'toronto_sun']
+    exclude = ['craigslist', 'iht', 'outlook_india', 'toronto_sun']
    d = os.path.dirname
    base = os.path.join(d(d(d(d(d(d(os.path.abspath(__file__))))))), 'resources', 'recipes')
    for x in os.walk(base):
@ -65,9 +65,9 @@ def serialize_builtin_recipes():
    from calibre.web.feeds.recipes import compile_recipe
    recipe_mapping = {}
    for rid, f in iterate_over_builtin_recipe_files():
-            recipe_class = compile_recipe(open(f, 'rb').read())
-            if recipe_class is not None:
-                recipe_mapping['builtin:'+rid] = recipe_class
+        recipe_class = compile_recipe(open(f, 'rb').read())
+        if recipe_class is not None:
+            recipe_mapping['builtin:'+rid] = recipe_class

    return serialize_collection(recipe_mapping)

--- a/src/calibre/web/fetch/simple.py
+++ b/src/calibre/web/fetch/simple.py
@ -188,11 +188,12 @@ class RecursiveFetcher(object):
        delta = time.time() - self.last_fetch_at
        if  delta < self.delay:
            time.sleep(delta)
-        if re.search(r'\s+|,', url) is not None:
-            purl = list(urlparse.urlparse(url))
-            for i in range(2, 6):
-                purl[i] = quote(purl[i])
-            url = urlparse.urlunparse(purl)
+        if isinstance(url, unicode):
+            url = url.encode('utf-8')
+        purl = list(urlparse.urlparse(url))
+        for i in range(2, 6):
+            purl[i] = quote(purl[i])
+        url = urlparse.urlunparse(purl)
        try:
            open_func = getattr(self.browser, 'open_novisit', self.browser.open)
            with closing(open_func(url, timeout=self.timeout)) as f:
--- a/src/cssutils/init.py
+++ b/src/cssutils/init.py
@ -17,7 +17,8 @@
    along with this program.  If not, see <http://www.gnu.org/licenses/>.


-A Python package to parse and build CSS Cascading Style Sheets. DOM only, not any rendering facilities!
+A Python package to parse and build CSS Cascading Style Sheets. DOM only, not
+any rendering facilities!

 Based upon and partly implementing the following specifications :

@ -26,30 +27,47 @@ Based upon and partly implementing the following specifications :
 `CSS 2.1 Errata  <http://www.w3.org/Style/css2-updates/CR-CSS21-20070719-errata.html>`__
    A few errata, mainly the definition of CHARSET_SYM tokens
 `CSS3 Module: Syntax <http://www.w3.org/TR/css3-syntax/>`__
-    Used in parts since cssutils 0.9.4. cssutils tries to use the features from CSS 2.1 and CSS 3 with preference to CSS3 but as this is not final yet some parts are from CSS 2.1
+    Used in parts since cssutils 0.9.4. cssutils tries to use the features from
+    CSS 2.1 and CSS 3 with preference to CSS3 but as this is not final yet some
+    parts are from CSS 2.1
 `MediaQueries <http://www.w3.org/TR/css3-mediaqueries/>`__
-    MediaQueries are part of ``stylesheets.MediaList`` since v0.9.4, used in @import and @media rules.
+    MediaQueries are part of ``stylesheets.MediaList`` since v0.9.4, used in
+    @import and @media rules.
 `Namespaces <http://dev.w3.org/csswg/css3-namespace/>`__
-    Added in v0.9.1, updated to definition in CSSOM in v0.9.4, updated in 0.9.5 for dev version
+    Added in v0.9.1, updated to definition in CSSOM in v0.9.4, updated in 0.9.5
+    for dev version
 `Selectors <http://www.w3.org/TR/css3-selectors/>`__
-    The selector syntax defined here (and not in CSS 2.1) should be parsable with cssutils (*should* mind though ;) )
+    The selector syntax defined here (and not in CSS 2.1) should be parsable
+    with cssutils (*should* mind though ;) )

 `DOM Level 2 Style CSS <http://www.w3.org/TR/DOM-Level-2-Style/css.html>`__
    DOM for package css
 `DOM Level 2 Style Stylesheets <http://www.w3.org/TR/DOM-Level-2-Style/stylesheets.html>`__
    DOM for package stylesheets
 `CSSOM <http://dev.w3.org/csswg/cssom/>`__
-    A few details (mainly the NamespaceRule DOM) is taken from here. Plan is to move implementation to the stuff defined here which is newer but still no REC so might change anytime...
+    A few details (mainly the NamespaceRule DOM) is taken from here. Plan is
+    to move implementation to the stuff defined here which is newer but still
+    no REC so might change anytime...


-The cssutils tokenizer is a customized implementation of `CSS3 Module: Syntax (W3C Working Draft 13 August 2003) <http://www.w3.org/TR/css3-syntax/>`__ which itself is based on the CSS 2.1 tokenizer. It tries to be as compliant as possible but uses some (helpful) parts of the CSS 2.1 tokenizer.
+The cssutils tokenizer is a customized implementation of `CSS3 Module: Syntax
+(W3C Working Draft 13 August 2003) <http://www.w3.org/TR/css3-syntax/>`__ which
+itself is based on the CSS 2.1 tokenizer. It tries to be as compliant as
+possible but uses some (helpful) parts of the CSS 2.1 tokenizer.

-I guess cssutils is neither CSS 2.1 nor CSS 3 compliant but tries to at least be able to parse both grammars including some more real world cases (some CSS hacks are actually parsed and serialized). Both official grammars are not final nor bugfree but still feasible. cssutils aim is not to be fully compliant to any CSS specification (the specifications seem to be in a constant flow anyway) but cssutils *should* be able to read and write as many as possible CSS stylesheets "in the wild" while at the same time implement the official APIs which are well documented. Some minor extensions are provided as well.
+I guess cssutils is neither CSS 2.1 nor CSS 3 compliant but tries to at least
+be able to parse both grammars including some more real world cases (some CSS
+hacks are actually parsed and serialized). Both official grammars are not final
+nor bugfree but still feasible. cssutils aim is not to be fully compliant to
+any CSS specification (the specifications seem to be in a constant flow anyway)
+but cssutils *should* be able to read and write as many as possible CSS
+stylesheets "in the wild" while at the same time implement the official APIs
+which are well documented. Some minor extensions are provided as well.

 Please visit http://cthedot.de/cssutils/ for more details.


-Tested with Python 2.5 on Windows Vista mainly.
+Tested with Python 2.6 on Windows 7 mainly.


 This library may be used ``from cssutils import *`` which
@ -70,13 +88,16 @@ Usage may be::
 __all__ = ['css', 'stylesheets', 'CSSParser', 'CSSSerializer']
 __docformat__ = 'restructuredtext'
 __author__ = 'Christof Hoeke with contributions by Walter Doerwald'
-__date__ = '$LastChangedDate:: 2009-10-17 15:12:28 -0600 #$:'
+__date__ = '$LastChangedDate:: 2009-11-26 16:31:32 -0700 #$:'

 VERSION = '0.9.7a1'

-__version__ = '%s $Id: __init__.py 1877 2009-10-17 21:12:28Z cthedot $' % VERSION
+__version__ = '%s $Id: __init__.py 1892 2009-11-26 23:31:32Z cthedot $' % VERSION

 import codec
+import os.path
+import urllib
+import urlparse
 import xml.dom

 # order of imports is important (partly circular)
@ -230,19 +251,22 @@ def getUrls(sheet):
                if u is not None:
                    yield u

-def replaceUrls(sheet, replacer):
+def replaceUrls(sheet, replacer, ignoreImportRules=False):
    """Replace all URLs in :class:`cssutils.css.CSSImportRule` or
    :class:`cssutils.css.CSSValue` objects of given `sheet`.

    :param sheet:
        :class:`cssutils.css.CSSStyleSheet` which is changed
    :param replacer:
-        a function which is called with a single argument `urlstring` which is
-        the current value of each url() excluding ``url(`` and ``)`` and
+        a function which is called with a single argument `urlstring` which 
+        is the current value of each url() excluding ``url(`` and ``)`` and
        surrounding single or double quotes.
+    :param ignoreImportRules:
+        if ``True`` does not call `replacer` with URLs from @import rules.
    """
-    for importrule in (r for r in sheet if r.type == r.IMPORT_RULE):
-        importrule.href = replacer(importrule.href)
+    if not ignoreImportRules:
+        for importrule in (r for r in sheet if r.type == r.IMPORT_RULE):
+            importrule.href = replacer(importrule.href)

    def setProperty(v):
        if v.CSS_PRIMITIVE_VALUE == v.cssValueType and\
@ -273,7 +297,7 @@ def resolveImports(sheet, target=None):
    @import rules which use media information are tried to be wrapped into
    @media rules so keeping the media information. This may not work in 
    all instances (if e.g. an @import rule itself contains an @import rule
-    with different media infos or if it is contains rules which may not be 
+    with different media infos or if it contains rules which may not be 
    used inside an @media block like @namespace rules.). In these cases
    the @import rule is kept as in the original sheet and a WARNING is issued.

@ -281,43 +305,110 @@ def resolveImports(sheet, target=None):
        in this given :class:`cssutils.css.CSSStyleSheet` all import rules are
        resolved and added to a resulting *flat* sheet.
    :param target:
-        A :class:`cssutils.css.CSSStyleSheet` object which will be the resulting
-        *flat* sheet if given
-    :returns: given `target` or a new :class:`cssutils.css.CSSStyleSheet` object
+        A :class:`cssutils.css.CSSStyleSheet` object which will be the
+        resulting *flat* sheet if given
+    :returns: given `target` or a new :class:`cssutils.css.CSSStyleSheet` 
+        object
    """
    if not target:
-        target = css.CSSStyleSheet()
+        target = css.CSSStyleSheet(href=sheet.href, 
+                                   media=sheet.media, 
+                                   title=sheet.title)
+
+    def getReplacer(targetbase):
+        "Return a replacer which uses base to return adjusted URLs"
+        basesch, baseloc, basepath, basequery, basefrag = urlparse.urlsplit(targetbase)
+        basepath, basepathfilename = os.path.split(basepath)
+
+        def replacer(url):
+            scheme, location, path, query, fragment = urlparse.urlsplit(url)
+            if not scheme and not location and not path.startswith(u'/'):
+                # relative
+                path, filename = os.path.split(path)
+                combined = os.path.normpath(os.path.join(basepath, path, filename))
+                return urllib.pathname2url(combined)
+            else:
+                # keep anything absolute
+                return url
+        
+        return replacer

-    #target.add(css.CSSComment(cssText=u'/* START %s */' % sheet.href))
    for rule in sheet.cssRules:
        if rule.type == rule.CHARSET_RULE:
            pass
        elif rule.type == rule.IMPORT_RULE:
            log.info(u'Processing @import %r' % rule.href, neverraise=True)
+            
            if rule.styleSheet:
-                target.add(css.CSSComment(cssText=u'/* START @import "%s" */' % rule.href))
-                if rule.media.mediaText == 'all':
-                    t = target
-                else:
-                    log.info(u'Replacing @import media with @media: %s' % 
-                             rule.media.mediaText, neverraise=True)
-                    t = css.CSSMediaRule(rule.media.mediaText)
+                # add all rules of @import to current sheet        
+                target.add(css.CSSComment(cssText=u'/* START @import "%s" */'
+                                          % rule.href))
+
                try:
-                    resolveImports(rule.styleSheet, t)
+                    # nested imports
+                    importedSheet = resolveImports(rule.styleSheet)
                except xml.dom.HierarchyRequestErr, e:
-                    log.warn(u'Cannot resolve @import: %s' % 
-                             e, neverraise=True)
+                    log.warn(u'@import: Cannot resolve target, keeping rule: %s'
+                             % e, neverraise=True)
                    target.add(rule)
                else:
-                    if t != target:
-                        target.add(t)
-                    t.add(css.CSSComment(cssText=u'/* END "%s" */' % rule.href))
+                    # adjust relative URI references 
+                    log.info(u'@import: Adjusting paths for %r' % rule.href, 
+                             neverraise=True)
+                    replaceUrls(importedSheet, 
+                                getReplacer(rule.href), 
+                                ignoreImportRules=True)
+                    
+                    # might have to wrap rules in @media if media given
+                    if rule.media.mediaText == u'all':
+                        mediaproxy = None
+                    else:
+                        keepimport = False
+                        for r in importedSheet:
+                            # check if rules present which may not be 
+                            # combined with media
+                            if r.type not in (r.COMMENT, 
+                                              r.STYLE_RULE, 
+                                              r.IMPORT_RULE):
+                                keepimport = True
+                                break
+                        if keepimport:
+                            log.warn(u'Cannot combine imported sheet with'
+                                     u' given media as other rules then'
+                                     u' comments or stylerules found %r,'
+                                     u' keeping %r' % (r,
+                                                       rule.cssText), 
+                                     neverraise=True)
+                            target.add(rule)
+                            continue
+                             
+                        # wrap in @media if media is not `all`
+                        log.info(u'@import: Wrapping some rules in @media '
+                                 u' to keep media: %s' 
+                                 % rule.media.mediaText, neverraise=True)
+                        mediaproxy = css.CSSMediaRule(rule.media.mediaText)
+                        
+                    for r in importedSheet:
+                        if mediaproxy:
+                            mediaproxy.add(r)
+                        else:
+                            # add to top sheet directly but are difficult anyway
+                            target.add(r)
+                            
+                    if mediaproxy:
+                        target.add(mediaproxy)
+                        
            else:
-                log.error(u'Cannot get referenced stylesheet %r' %
-                          rule.href, neverraise=True)
+                # keep @import as it is
+                log.error(u'Cannot get referenced stylesheet %r, keeping rule'
+                          % rule.href, neverraise=True)
                target.add(rule)
+                
+                
+
        else:
            target.add(rule)
+    
    return target


--- a/src/cssutils/css/cssstyledeclaration.py
+++ b/src/cssutils/css/cssstyledeclaration.py
@ -51,7 +51,7 @@ TODO:
 """
 __all__ = ['CSSStyleDeclaration', 'Property']
 __docformat__ = 'restructuredtext'
-__version__ = '$Id: cssstyledeclaration.py 1870 2009-10-17 19:56:59Z cthedot $'
+__version__ = '$Id: cssstyledeclaration.py 1879 2009-11-17 20:35:04Z cthedot $'

 from cssproperties import CSS2Properties
 from property import Property
@ -581,6 +581,9 @@ class CSSStyleDeclaration(CSS2Properties, cssutils.util.Base2):
        if isinstance(name, Property):
            newp = name 
            name = newp.literalname
+        elif not value:
+            # empty string or None effectively removed property
+            return self.removeProperty(name)
        else:
            newp = Property(name, value, priority)
        if not newp.wellformed:
--- a/src/cssutils/css/cssunknownrule.py
+++ b/src/cssutils/css/cssunknownrule.py
@ -1,7 +1,7 @@
 """CSSUnknownRule implements DOM Level 2 CSS CSSUnknownRule."""
 __all__ = ['CSSUnknownRule']
 __docformat__ = 'restructuredtext'
-__version__ = '$Id: cssunknownrule.py 1638 2009-01-13 20:39:33Z cthedot $'
+__version__ = '$Id: cssunknownrule.py 1897 2009-12-17 22:09:06Z cthedot $'

 import cssrule
 import cssutils
@ -98,6 +98,20 @@ class CSSUnknownRule(cssrule.CSSRule):
                        token=token)
                    return expected

+            def FUNCTION(expected, seq, token, tokenizer=None):
+                # handled as opening (
+                type_, val, line, col = token
+                val = self._tokenvalue(token)
+                if expected != 'EOF':
+                    new['nesting'].append(u'(')
+                    seq.append(val, type_, line=line, col=col)
+                    return expected
+                else:
+                    new['wellformed'] = False
+                    self._log.error(u'CSSUnknownRule: Expected end of rule.',
+                        token=token)
+                    return expected                
+
            def EOF(expected, seq, token, tokenizer=None):
                "close all blocks and return 'EOF'"
                for x in reversed(new['nesting']):
@ -154,6 +168,7 @@ class CSSUnknownRule(cssrule.CSSRule):
                seq=newseq, tokenizer=tokenizer,
                productions={'CHAR': CHAR,
                             'EOF': EOF,
+                             'FUNCTION': FUNCTION,
                             'INVALID': INVALID,
                             'STRING': STRING,
                             'URI': URI,
--- a/src/cssutils/css/property.py
+++ b/src/cssutils/css/property.py
@ -1,7 +1,7 @@
 """Property is a single CSS property in a CSSStyleDeclaration."""
 __all__ = ['Property']
 __docformat__ = 'restructuredtext'
-__version__ = '$Id: property.py 1868 2009-10-17 19:36:54Z cthedot $'
+__version__ = '$Id: property.py 1878 2009-11-17 20:16:26Z cthedot $'

 from cssutils.helper import Deprecated
 from cssvalue import CSSValue
@ -68,13 +68,10 @@ class Property(cssutils.util.Base):
        self.__nametoken = None
        self._name = u''
        self._literalname = u''
+        self.seqs[1] = CSSValue(parent=self)
        if name:
            self.name = name
-
-        if value:
            self.cssValue = value
-        else:
-            self.seqs[1] = CSSValue(parent=self)

        self._priority = u''
        self._literalpriority = u''
@ -246,14 +243,14 @@ class Property(cssutils.util.Base):
        if self._mediaQuery and not cssText:
            self.seqs[1] = CSSValue(parent=self)
        else:
-            #if not self.seqs[1]:
-            #    self.seqs[1] = CSSValue(parent=self)
+            oldvalue = self.seqs[1].cssText
+            try:
+                self.seqs[1].cssText = cssText
+            except:
+                self.seqs[1].cssText = oldvalue
+                raise
            
-            self.seqs[1] = CSSValue(parent=self)
-            
-            self.seqs[1].cssText = cssText
            self.wellformed = self.wellformed and self.seqs[1].wellformed
-            # self.valid = self.valid and self.cssValue.valid

    cssValue = property(_getCSSValue, _setCSSValue,
        doc="(cssutils) CSSValue object of this property")
--- a/src/cssutils/sac.py
+++ b/src/cssutils/sac.py
@ -0,0 +1,428 @@
+#!/usr/bin/env python
+"""A validating CSSParser"""
+__all__ = ['CSSParser']
+__docformat__ = 'restructuredtext'
+__version__ = '$Id: parse.py 1754 2009-05-30 14:50:13Z cthedot $'
+
+import helper
+import codecs
+import errorhandler
+import os
+import tokenize2
+import urllib
+import sys
+ 
+    
+class ErrorHandler(object):
+    """Basic class for CSS error handlers.
+    
+    This class class provides a default implementation ignoring warnings and
+    recoverable errors and throwing a SAXParseException for fatal errors.
+
+    If a CSS application needs to implement customized error handling, it must
+    extend this class and then register an instance with the CSS parser
+    using the parser's setErrorHandler method. The parser will then report all
+    errors and warnings through this interface.
+
+    The parser shall use this class instead of throwing an exception: it is
+    up to the application whether to throw an exception for different types of
+    errors and warnings. Note, however, that there is no requirement that the
+    parser continue to provide useful information after a call to fatalError
+    (in other words, a CSS driver class could catch an exception and report a 
+    fatalError).
+    """
+    def __init__(self):
+        self._log = errorhandler.ErrorHandler()
+    
+    def error(self, exception, token=None):
+        self._log.error(exception, token, neverraise=True)
+
+    def fatal(self, exception, token=None):
+        self._log.fatal(exception, token)
+
+    def warn(self, exception, token=None):
+        self._log.warn(exception, token, neverraise=True)
+
+     
+class DocumentHandler(object):
+    """
+     void     endFontFace()
+              Receive notification of the end of a font face statement.
+     void     endMedia(SACMediaList media)
+              Receive notification of the end of a media statement.
+     void     endPage(java.lang.String name, java.lang.String pseudo_page)
+              Receive notification of the end of a media statement.
+     void     importStyle(java.lang.String uri, SACMediaList media, java.lang.String defaultNamespaceURI)
+              Receive notification of a import statement in the style sheet.
+     void     startFontFace()
+              Receive notification of the beginning of a font face statement.
+     void     startMedia(SACMediaList media)
+              Receive notification of the beginning of a media statement.
+     void     startPage(java.lang.String name, java.lang.String pseudo_page)
+              Receive notification of the beginning of a page statement.
+    """
+    def __init__(self):
+        def log(msg):
+            sys.stderr.write('INFO\t%s\n' % msg)
+        self._log = log
+    
+    def comment(self, text, line=None, col=None):
+        "Receive notification of a comment."
+        self._log("comment %r at [%s, %s]" % (text, line, col))
+    
+    def startDocument(self, encoding):
+        "Receive notification of the beginning of a style sheet."
+        # source
+        self._log("startDocument encoding=%s" % encoding)
+
+    def endDocument(self, source=None, line=None, col=None):
+        "Receive notification of the end of a document."
+        self._log("endDocument EOF")
+    
+    def importStyle(self, uri, media, name, line=None, col=None):
+        "Receive notification of a import statement in the style sheet."
+        # defaultNamespaceURI???
+        self._log("importStyle at [%s, %s]" % (line, col))
+    
+    def namespaceDeclaration(self, prefix, uri, line=None, col=None):
+        "Receive notification of an unknown rule t-rule not supported by this parser."
+        # prefix might be None!
+        self._log("namespaceDeclaration at [%s, %s]" % (line, col))
+    
+    def startSelector(self, selectors=None, line=None, col=None):
+        "Receive notification of the beginning of a rule statement."
+        # TODO selectorList!
+        self._log("startSelector at [%s, %s]" % (line, col))
+
+    def endSelector(self, selectors=None, line=None, col=None):
+        "Receive notification of the end of a rule statement."
+        self._log("endSelector at [%s, %s]" % (line, col))
+    
+    def property(self, name, value='TODO', important=False, line=None, col=None):
+        "Receive notification of a declaration."
+        # TODO: value is LexicalValue?
+        self._log("property %r at [%s, %s]" % (name, line, col))
+
+    def ignorableAtRule(self, atRule, line=None, col=None):
+        "Receive notification of an unknown rule t-rule not supported by this parser."
+        self._log("ignorableAtRule %r at [%s, %s]" % (atRule, line, col))
+        
+
+
+class EchoHandler(DocumentHandler):
+    "Echos all input to property `out`"
+    def __init__(self):
+        super(EchoHandler, self).__init__()
+        self._out = []
+        
+    out = property(lambda self: u''.join(self._out))
+
+    def startDocument(self, encoding):
+        super(EchoHandler, self).startDocument(encoding)
+        if u'utf-8' != encoding:
+            self._out.append(u'@charset "%s";\n' % encoding)
+    
+#    def comment(self, text, line=None, col=None):
+#        self._out.append(u'/*%s*/' % text)
+
+    def importStyle(self, uri, media, name, line=None, col=None):
+        "Receive notification of a import statement in the style sheet."
+        # defaultNamespaceURI???
+        super(EchoHandler, self).importStyle(uri, media, name, line, col)
+        self._out.append(u'@import %s%s%s;\n' % (helper.string(uri),
+                                               u'%s ' % media if media else u'',
+                                               u'%s ' % name if name else u'')
+        )
+        
+        
+    def namespaceDeclaration(self, prefix, uri, line=None, col=None):
+        super(EchoHandler, self).namespaceDeclaration(prefix, uri, line, col)
+        self._out.append(u'@namespace %s%s;\n' % (u'%s ' % prefix if prefix else u'', 
+                                                helper.string(uri)))
+
+    def startSelector(self, selectors=None, line=None, col=None):
+        super(EchoHandler, self).startSelector(selectors, line, col)
+        if selectors:
+            self._out.append(u', '.join(selectors))
+        self._out.append(u' {\n')
+
+    def endSelector(self, selectors=None, line=None, col=None):
+        self._out.append(u'    }')
+        
+    def property(self, name, value, important=False, line=None, col=None):
+        super(EchoHandler, self).property(name, value,  line, col)
+        self._out.append(u'    %s: %s%s;\n' % (name, value, 
+                                        u' !important' if important else u''))
+  
+
+class Parser(object):
+    """
+    java.lang.String     getParserVersion()
+        Returns a string about which CSS language is supported by this parser.
+     boolean     parsePriority(InputSource source)
+          Parse a CSS priority value (e.g.
+     LexicalUnit     parsePropertyValue(InputSource source)
+          Parse a CSS property value.
+     void     parseRule(InputSource source)
+          Parse a CSS rule.
+     SelectorList     parseSelectors(InputSource source)
+          Parse a comma separated list of selectors.
+     void     parseStyleDeclaration(InputSource source)
+          Parse a CSS style declaration (without '{' and '}').
+     void     parseStyleSheet(InputSource source)
+          Parse a CSS document.
+     void     parseStyleSheet(java.lang.String uri)
+          Parse a CSS document from a URI.
+     void     setConditionFactory(ConditionFactory conditionFactory)
+           
+     void     setDocumentHandler(DocumentHandler handler)
+          Allow an application to register a document event handler.
+     void     setErrorHandler(ErrorHandler handler)
+          Allow an application to register an error event handler.
+     void     setLocale(java.util.Locale locale)
+          Allow an application to request a locale for errors and warnings.
+     void     setSelectorFactory(SelectorFactory selectorFactory) 
+    """
+    def __init__(self, documentHandler=None, errorHandler=None):
+        self._tokenizer = tokenize2.Tokenizer()
+        if documentHandler:
+            self.setDocumentHandler(documentHandler)
+        else:
+            self.setDocumentHandler(DocumentHandler())
+
+        if errorHandler:
+            self.setErrorHandler(errorHandler)
+        else:
+            self.setErrorHandler(ErrorHandler())
+    
+    def parseString(self, cssText, encoding=None):
+        if isinstance(cssText, str):
+            cssText = codecs.getdecoder('css')(cssText, encoding=encoding)[0]
+        
+        tokens = self._tokenizer.tokenize(cssText, fullsheet=True)
+                
+        def COMMENT(val, line, col):
+            self._handler.comment(val[2:-2], line, col)
+
+        def EOF(val, line, col):
+            self._handler.endDocument(val, line, col)
+                
+        def simple(t):            
+            map = {'COMMENT': COMMENT,
+                   'S': lambda val, line, col: None,
+                   'EOF': EOF}
+            type_, val, line, col = t
+            if type_ in map:
+                map[type_](val, line, col)
+                return True
+            else:
+                return False
+                        
+        # START PARSING
+        t = tokens.next()
+        type_, val, line, col = t
+        
+        encoding = 'utf-8'                
+        if 'CHARSET_SYM' == type_:
+            # @charset "encoding";
+            # S
+            encodingtoken = tokens.next()
+            semicolontoken = tokens.next()
+            if 'STRING' == type_:
+                encoding = helper.stringvalue(val)
+            # ;
+            if 'STRING' == encodingtoken[0] and semicolontoken:
+                encoding = helper.stringvalue(encodingtoken[1])    
+            else:
+                self._errorHandler.fatal(u'Invalid @charset')
+                
+            t = tokens.next()
+            type_, val, line, col = t
+            
+        self._handler.startDocument(encoding)
+                
+        while True:
+            start = (line, col)
+            try:
+                if simple(t):
+                    pass
+
+                elif 'ATKEYWORD' == type_ or type_ in ('PAGE_SYM', 'MEDIA_SYM', 'FONT_FACE_SYM'):
+                    atRule = [val]
+                    braces = 0
+                    while True:
+                        # read till end ; 
+                        # TODO: or {}
+                        t = tokens.next()
+                        type_, val, line, col = t
+                        atRule.append(val) 
+                        if u';' == val and not braces:
+                            break
+                        elif u'{' == val:
+                            braces += 1
+                        elif u'}' == val:
+                            braces -= 1
+                            if braces == 0:
+                                break
+                            
+                    self._handler.ignorableAtRule(u''.join(atRule), *start)
+
+                elif 'IMPORT_SYM' == type_:
+                    # import URI or STRING media? name?
+                    uri, media, name = None, None, None
+                    while True:
+                        t = tokens.next()
+                        type_, val, line, col = t
+                        if 'STRING' == type_:
+                            uri = helper.stringvalue(val)
+                        elif 'URI' == type_:
+                            uri = helper.urivalue(val)
+                        elif u';' == val:
+                            break
+                    
+                    if uri:    
+                        self._handler.importStyle(uri, media, name)
+                    else:
+                        self._errorHandler.error(u'Invalid @import'
+                                                 u' declaration at %r' 
+                                                 % (start,))
+                        
+                elif 'NAMESPACE_SYM' == type_:
+                    prefix, uri = None, None
+                    while True:
+                        t = tokens.next()
+                        type_, val, line, col = t
+                        if 'IDENT' == type_:
+                            prefix = val
+                        elif 'STRING' == type_:
+                            uri = helper.stringvalue(val)
+                        elif 'URI' == type_:
+                            uri = helper.urivalue(val)
+                        elif u';' == val:
+                            break
+                    if uri:
+                        self._handler.namespaceDeclaration(prefix, uri, *start)
+                    else:
+                        self._errorHandler.error(u'Invalid @namespace'
+                                                 u' declaration at %r' 
+                                                 % (start,))
+    
+                else:
+                    # CSSSTYLERULE
+                    selector = []
+                    selectors = []
+                    while True:
+                        # selectors[, selector]* {
+                        if 'S' == type_:
+                            selector.append(u' ')
+                        elif simple(t):
+                            pass
+                        elif u',' == val:
+                            selectors.append(u''.join(selector).strip())
+                            selector = []
+                        elif u'{' == val:
+                            selectors.append(u''.join(selector).strip())
+                            self._handler.startSelector(selectors, *start)
+                            break
+                        else:
+                            selector.append(val)
+                            
+                        t = tokens.next()
+                        type_, val, line, col = t
+                                            
+                    end = None
+                    while True:
+                        # name: value [!important][;name: value [!important]]*;?
+                        name, value, important = None, [], False
+                        
+                        while True:
+                            # name:
+                            t = tokens.next()
+                            type_, val, line, col = t
+                            if 'S' == type_:
+                                pass
+                            elif simple(t):
+                                pass
+                            elif 'IDENT' == type_:
+                                if name:
+                                    self._errorHandler.error('more than one property name', t)
+                                else:
+                                    name = val
+                            elif u':' == val:
+                                if not name:
+                                    self._errorHandler.error('no property name', t)
+                                break
+                            elif u';' == val:
+                                self._errorHandler.error('premature end of property', t)
+                                end = val
+                                break
+                            elif u'}' == val:
+                                if name:
+                                    self._errorHandler.error('premature end of property', t)
+                                end = val
+                                break
+                            else:
+                                self._errorHandler.error('unexpected property name token %r' % val, t)
+
+                        while not u';' == end and not u'}' == end:
+                            # value !;}
+                            t = tokens.next()
+                            type_, val, line, col = t
+                            
+                            if 'S' == type_:
+                                value.append(u' ')
+                            elif simple(t):
+                                pass
+                            elif u'!' == val or u';' == val or u'}' == val:
+                                value = ''.join(value).strip()
+                                if not value:
+                                    self._errorHandler.error('premature end of property (no value)', t)
+                                end = val
+                                break
+                            else:
+                                value.append(val)
+
+                        while u'!' == end:
+                            # !important
+                            t = tokens.next()
+                            type_, val, line, col = t
+                            
+                            if simple(t):
+                                pass
+                            elif u'IDENT' == type_ and not important:
+                                important = True
+                            elif u';' == val or u'}' == val:
+                                end = val
+                                break
+                            else:
+                                self._errorHandler.error('unexpected priority token %r' % val)
+
+                        if name and value:
+                            self._handler.property(name, value, important)
+                            
+                        if u'}' == end:
+                            self._handler.endSelector(selectors, line=line, col=col)
+                            break
+                        else:
+                            # reset
+                            end = None
+
+                    else:
+                        self._handler.endSelector(selectors, line=line, col=col)
+
+                t = tokens.next()
+                type_, val, line, col = t
+
+            except StopIteration:
+                break
+            
+
+    
+    def setDocumentHandler(self, handler):
+        "Allow an application to register a document event `handler`."
+        self._handler = handler
+        
+    def setErrorHandler(self, handler):
+        "TODO"
+        self._errorHandler = handler
+        
--- a/src/cssutils/script.py
+++ b/src/cssutils/script.py
@ -318,6 +318,8 @@ def csscombine(path=None, url=None,
        `path` or `url`
            path  or URL to a CSSStyleSheet which imports other sheets which
            are then combined into one sheet
+        `sourceencoding`
+            explicit encoding of the source proxysheet, default 'utf-8'
        `targetencoding`
            encoding of the combined stylesheet, default 'utf-8'
        `minify`
@ -350,4 +352,3 @@ def csscombine(path=None, url=None,
        cssText = result.cssText
    
    return cssText
-
--- a/src/cssutils/serialize.py
+++ b/src/cssutils/serialize.py
@ -3,7 +3,7 @@
 """cssutils serializer"""
 __all__ = ['CSSSerializer', 'Preferences']
 __docformat__ = 'restructuredtext'
-__version__ = '$Id: serialize.py 1872 2009-10-17 21:00:40Z cthedot $'
+__version__ = '$Id: serialize.py 1898 2009-12-19 12:17:04Z cthedot $'

 import codecs
 import cssutils
@ -191,7 +191,6 @@ class Out(object):
            add ``*spacer`` except ``space=False``
        """
        prefspace = self.ser.prefs.spacer
-        
        if val or typ in ('STRING', 'URI'):
            # PRE
            if 'COMMENT' == typ:
@ -230,7 +229,10 @@ class Out(object):
            if indent:
                self.out.append(self.ser._indentblock(val, self.ser._level+1))
            else:
+                if val.endswith(u' '):
+                    self._remove_last_if_S()
                self.out.append(val)
+                        
            # POST
            if lineSeparator:
                # Property , ...
@ -238,6 +240,9 @@ class Out(object):
            elif val in u'+>~': # enclose selector combinator
                self.out.insert(-1, self.ser.prefs.selectorCombinatorSpacer)
                self.out.append(self.ser.prefs.selectorCombinatorSpacer)
+            elif u')' == val and not keepS: # CHAR funcend 
+                # TODO: pref?
+                self.out.append(u' ')
            elif u',' == val: # list
                self.out.append(self.ser.prefs.listItemSpacer)
            elif u':' == val: # prop